Add support for IQ1_S, IQ3_S, IQ2_S, IQ4_XS. IQ4_NL

Co-authored-by: ManniX-ITA <20623405+mann1x@users.noreply.github.com>
s/DisplayLongest/String/
2024-05-03 14:51:07 -07:00 · 2024-05-03 13:18:28 -07:00 · 2024-05-03 13:18:28 -07:00 · 2024-05-03 13:18:28 -07:00 · 2024-05-03 13:18:28 -07:00 · 2024-05-03 13:18:28 -07:00
41 changed files with 1739 additions and 1102 deletions
--- a/cmd/cmd.go
+++ b/cmd/cmd.go
@@ -34,7 +34,6 @@ import (
 	"github.com/ollama/ollama/api"
 	"github.com/ollama/ollama/auth"
 	"github.com/ollama/ollama/format"
-	"github.com/ollama/ollama/parser"
 	"github.com/ollama/ollama/progress"
 	"github.com/ollama/ollama/server"
 	"github.com/ollama/ollama/types/errtypes"
@@ -57,12 +56,13 @@ func CreateHandler(cmd *cobra.Command, args []string) error {
 	p := progress.NewProgress(os.Stderr)
 	defer p.Stop()

-	modelfile, err := os.ReadFile(filename)
+	f, err := os.Open(filename)
 	if err != nil {
 		return err
 	}
+	defer f.Close()

-	commands, err := parser.Parse(bytes.NewReader(modelfile))
+	modelfile, err := model.ParseFile(f)
 	if err != nil {
 		return err
 	}
@@ -76,10 +76,10 @@ func CreateHandler(cmd *cobra.Command, args []string) error {
 	spinner := progress.NewSpinner(status)
 	p.Add(status, spinner)

-	for _, c := range commands {
-		switch c.Name {
+	for i := range modelfile.Commands {
+		switch modelfile.Commands[i].Name {
 		case "model", "adapter":
-			path := c.Args
+			path := modelfile.Commands[i].Args
 			if path == "~" {
 				path = home
 			} else if strings.HasPrefix(path, "~/") {
@@ -91,7 +91,7 @@ func CreateHandler(cmd *cobra.Command, args []string) error {
 			}

 			fi, err := os.Stat(path)
-			if errors.Is(err, os.ErrNotExist) && c.Name == "model" {
+			if errors.Is(err, os.ErrNotExist) && modelfile.Commands[i].Name == "model" {
 				continue
 			} else if err != nil {
 				return err
@@ -114,13 +114,7 @@ func CreateHandler(cmd *cobra.Command, args []string) error {
 				return err
 			}

-			name := c.Name
-			if c.Name == "model" {
-				name = "from"
-			}
-
-			re := regexp.MustCompile(fmt.Sprintf(`(?im)^(%s)\s+%s\s*$`, name, c.Args))
-			modelfile = re.ReplaceAll(modelfile, []byte("$1 @"+digest))
+			modelfile.Commands[i].Args = "@" + digest
 		}
 	}

@@ -150,7 +144,7 @@ func CreateHandler(cmd *cobra.Command, args []string) error {

 	quantization, _ := cmd.Flags().GetString("quantization")

-	request := api.CreateRequest{Name: args[0], Modelfile: string(modelfile), Quantization: quantization}
+	request := api.CreateRequest{Name: args[0], Modelfile: modelfile.String(), Quantization: quantization}
 	if err := client.Create(cmd.Context(), &request, fn); err != nil {
 		return err
 	}
--- a/convert/convert.go
+++ b/convert/convert.go
@@ -5,6 +5,7 @@ import (
 	"encoding/binary"
 	"encoding/json"
 	"fmt"
+	"io"
 	"log/slog"
 	"os"
 	"path/filepath"
@@ -47,7 +48,7 @@ type ByteOrder interface {
 type ModelArch interface {
 	GetTensors() error
 	LoadVocab() error
-	WriteGGUF() (string, error)
+	WriteGGUF(io.WriteSeeker) error
 }

 type ModelFormat interface {
--- a/convert/gemma.go
+++ b/convert/gemma.go
@@ -94,7 +94,7 @@ func (m *GemmaModel) LoadVocab() error {
 	return nil
 }

-func (m *GemmaModel) WriteGGUF() (string, error) {
+func (m *GemmaModel) WriteGGUF(ws io.WriteSeeker) error {
 	kv := llm.KV{
 		"general.architecture":                   "gemma",
 		"general.name":                           m.Name,
@@ -122,16 +122,5 @@ func (m *GemmaModel) WriteGGUF() (string, error) {
 		"tokenizer.ggml.add_eos_token":    false,
 	}

-	f, err := os.CreateTemp("", "ollama-gguf")
-	if err != nil {
-		return "", err
-	}
-	defer f.Close()
-
-	mod := llm.NewGGUFV3(m.Params.ByteOrder)
-	if err := mod.Encode(f, kv, m.Tensors); err != nil {
-		return "", err
-	}
-
-	return f.Name(), nil
+	return llm.NewGGUFV3(m.Params.ByteOrder).Encode(ws, kv, m.Tensors)
 }
--- a/convert/llama.go
+++ b/convert/llama.go
@@ -132,7 +132,7 @@ func (m *LlamaModel) LoadVocab() error {
 	return nil
 }

-func (m *LlamaModel) WriteGGUF() (string, error) {
+func (m *LlamaModel) WriteGGUF(ws io.WriteSeeker) error {
 	kv := llm.KV{
 		"general.architecture":                   "llama",
 		"general.name":                           m.Name,
@@ -161,16 +161,9 @@ func (m *LlamaModel) WriteGGUF() (string, error) {

 	f, err := os.CreateTemp("", "ollama-gguf")
 	if err != nil {
-		return "", err
+		return err
 	}
 	defer f.Close()

-	mod := llm.NewGGUFV3(m.Params.ByteOrder)
-	if err := mod.Encode(f, kv, m.Tensors); err != nil {
-		return "", err
-	}
-
-	slog.Debug(fmt.Sprintf("gguf file = %s", f.Name()))
-
-	return f.Name(), nil
+	return llm.NewGGUFV3(m.Params.ByteOrder).Encode(f, kv, m.Tensors)
 }
--- a/convert/mistral.go
+++ b/convert/mistral.go
@@ -132,7 +132,7 @@ func (m *MistralModel) LoadVocab() error {
 	return nil
 }

-func (m *MistralModel) WriteGGUF() (string, error) {
+func (m *MistralModel) WriteGGUF(ws io.WriteSeeker) error {
 	kv := llm.KV{
 		"general.architecture":                   "llama",
 		"general.name":                           m.Name,
@@ -158,16 +158,5 @@ func (m *MistralModel) WriteGGUF() (string, error) {
 		"tokenizer.ggml.unknown_token_id": uint32(0),
 	}

-	f, err := os.CreateTemp("", "ollama-gguf")
-	if err != nil {
-		return "", err
-	}
-	defer f.Close()
-
-	mod := llm.NewGGUFV3(m.Params.ByteOrder)
-	if err := mod.Encode(f, kv, m.Tensors); err != nil {
-		return "", err
-	}
-
-	return f.Name(), nil
+	return llm.NewGGUFV3(m.Params.ByteOrder).Encode(ws, kv, m.Tensors)
 }
--- a/convert/mixtral.go
+++ b/convert/mixtral.go
@@ -1,7 +1,7 @@
 package convert

 import (
-	"os"
+	"io"
 	"regexp"

 	"github.com/ollama/ollama/llm"
@@ -47,7 +47,7 @@ func (m *MixtralModel) LoadVocab() error {
 	return nil
 }

-func (m *MixtralModel) WriteGGUF() (string, error) {
+func (m *MixtralModel) WriteGGUF(ws io.WriteSeeker) error {
 	kv := llm.KV{
 		"general.architecture":          "llama",
 		"general.name":                  m.Name,
@@ -81,16 +81,5 @@ func (m *MixtralModel) WriteGGUF() (string, error) {
 		"tokenizer.ggml.add_eos_token":    false,
 	}

-	f, err := os.CreateTemp("", "ollama-gguf")
-	if err != nil {
-		return "", err
-	}
-	defer f.Close()
-
-	mod := llm.NewGGUFV3(m.Params.ByteOrder)
-	if err := mod.Encode(f, kv, m.Tensors); err != nil {
-		return "", err
-	}
-
-	return f.Name(), nil
+	return llm.NewGGUFV3(m.Params.ByteOrder).Encode(ws, kv, m.Tensors)
 }
--- a/docs/api.md
+++ b/docs/api.md
@@ -17,7 +17,7 @@

 ### Model names

-Model names follow a `model:tag` format, where `model` can have an optional namespace such as `example/model`. Some examples are `orca-mini:3b-q4_1` and `llama2:70b`. The tag is optional and, if not provided, will default to `latest`. The tag is used to identify a specific version.
+Model names follow a `model:tag` format, where `model` can have an optional namespace such as `example/model`. Some examples are `orca-mini:3b-q4_1` and `llama3:70b`. The tag is optional and, if not provided, will default to `latest`. The tag is used to identify a specific version.

 ### Durations

@@ -66,7 +66,7 @@ Enable JSON mode by setting the `format` parameter to `json`. This will structur

 ```shell
 curl http://localhost:11434/api/generate -d '{
-  "model": "llama2",
+  "model": "llama3",
  "prompt": "Why is the sky blue?"
 }'
 ```
@@ -77,7 +77,7 @@ A stream of JSON objects is returned:

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "response": "The",
  "done": false
@@ -99,7 +99,7 @@ To calculate how fast the response is generated in tokens per second (token/s),

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "response": "",
  "done": true,
@@ -121,7 +121,7 @@ A response can be received in one reply when streaming is off.

 ```shell
 curl http://localhost:11434/api/generate -d '{
-  "model": "llama2",
+  "model": "llama3",
  "prompt": "Why is the sky blue?",
  "stream": false
 }'
@@ -133,7 +133,7 @@ If `stream` is set to `false`, the response will be a single JSON object:

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "response": "The sky is blue because it is the color of the sky.",
  "done": true,
@@ -155,7 +155,7 @@ If `stream` is set to `false`, the response will be a single JSON object:

 ```shell
 curl http://localhost:11434/api/generate -d '{
-  "model": "llama2",
+  "model": "llama3",
  "prompt": "What color is the sky at different times of the day? Respond using JSON",
  "format": "json",
  "stream": false
@@ -166,7 +166,7 @@ curl http://localhost:11434/api/generate -d '{

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-11-09T21:07:55.186497Z",
  "response": "{\n\"morning\": {\n\"color\": \"blue\"\n},\n\"noon\": {\n\"color\": \"blue-gray\"\n},\n\"afternoon\": {\n\"color\": \"warm gray\"\n},\n\"evening\": {\n\"color\": \"orange\"\n}\n}\n",
  "done": true,
@@ -289,7 +289,7 @@ If you want to set custom options for the model at runtime rather than in the Mo

 ```shell
 curl http://localhost:11434/api/generate -d '{
-  "model": "llama2",
+  "model": "llama3",
  "prompt": "Why is the sky blue?",
  "stream": false,
  "options": {
@@ -332,7 +332,7 @@ curl http://localhost:11434/api/generate -d '{

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "response": "The sky is blue because it is the color of the sky.",
  "done": true,
@@ -354,7 +354,7 @@ If an empty prompt is provided, the model will be loaded into memory.

 ```shell
 curl http://localhost:11434/api/generate -d '{
-  "model": "llama2"
+  "model": "llama3"
 }'
 ```

@@ -364,7 +364,7 @@ A single JSON object is returned:

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-12-18T19:52:07.071755Z",
  "response": "",
  "done": true
@@ -407,7 +407,7 @@ Send a chat message with a streaming response.

 ```shell
 curl http://localhost:11434/api/chat -d '{
-  "model": "llama2",
+  "model": "llama3",
  "messages": [
    {
      "role": "user",
@@ -423,7 +423,7 @@ A stream of JSON objects is returned:

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "message": {
    "role": "assistant",
@@ -438,7 +438,7 @@ Final response:

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "done": true,
  "total_duration": 4883583458,
@@ -456,7 +456,7 @@ Final response:

 ```shell
 curl http://localhost:11434/api/chat -d '{
-  "model": "llama2",
+  "model": "llama3",
  "messages": [
    {
      "role": "user",
@@ -471,7 +471,7 @@ curl http://localhost:11434/api/chat -d '{

 ```json
 {
-  "model": "registry.ollama.ai/library/llama2:latest",
+  "model": "registry.ollama.ai/library/llama3:latest",
  "created_at": "2023-12-12T14:13:43.416799Z",
  "message": {
    "role": "assistant",
@@ -495,7 +495,7 @@ Send a chat message with a conversation history. You can use this same approach

 ```shell
 curl http://localhost:11434/api/chat -d '{
-  "model": "llama2",
+  "model": "llama3",
  "messages": [
    {
      "role": "user",
@@ -519,7 +519,7 @@ A stream of JSON objects is returned:

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-08-04T08:52:19.385406455-07:00",
  "message": {
    "role": "assistant",
@@ -533,7 +533,7 @@ Final response:

 ```json
 {
-  "model": "llama2",
+  "model": "llama3",
  "created_at": "2023-08-04T19:22:45.499127Z",
  "done": true,
  "total_duration": 8113331500,
@@ -591,7 +591,7 @@ curl http://localhost:11434/api/chat -d '{

 ```shell
 curl http://localhost:11434/api/chat -d '{
-  "model": "llama2",
+  "model": "llama3",
  "messages": [
    {
      "role": "user",
@@ -609,7 +609,7 @@ curl http://localhost:11434/api/chat -d '{

 ```json
 {
-  "model": "registry.ollama.ai/library/llama2:latest",
+  "model": "registry.ollama.ai/library/llama3:latest",
  "created_at": "2023-12-12T14:13:43.416799Z",
  "message": {
    "role": "assistant",
@@ -651,7 +651,7 @@ Create a new model from a `Modelfile`.
 ```shell
 curl http://localhost:11434/api/create -d '{
  "name": "mario",
-  "modelfile": "FROM llama2\nSYSTEM You are mario from Super Mario Bros."
+  "modelfile": "FROM llama3\nSYSTEM You are mario from Super Mario Bros."
 }'
 ```

@@ -758,7 +758,7 @@ A single JSON object will be returned.
      }
    },
    {
-      "name": "llama2:latest",
+      "name": "llama3:latest",
      "modified_at": "2023-12-07T09:32:18.757212583-08:00",
      "size": 3825819519,
      "digest": "fe938a131f40e6f6d40083c9f0f430a515233eb2edaa6d72eb85c50d64f2300e",
@@ -792,7 +792,7 @@ Show information about a model including details, modelfile, template, parameter

 ```shell
 curl http://localhost:11434/api/show -d '{
-  "name": "llama2"
+  "name": "llama3"
 }'
 ```

@@ -827,8 +827,8 @@ Copy a model. Creates a model with another name from an existing model.

 ```shell
 curl http://localhost:11434/api/copy -d '{
-  "source": "llama2",
-  "destination": "llama2-backup"
+  "source": "llama3",
+  "destination": "llama3-backup"
 }'
 ```

@@ -854,7 +854,7 @@ Delete a model and its data.

 ```shell
 curl -X DELETE http://localhost:11434/api/delete -d '{
-  "name": "llama2:13b"
+  "name": "llama3:13b"
 }'
 ```

@@ -882,7 +882,7 @@ Download a model from the ollama library. Cancelled pulls are resumed from where

 ```shell
 curl http://localhost:11434/api/pull -d '{
-  "name": "llama2"
+  "name": "llama3"
 }'
 ```

--- a/docs/faq.md
+++ b/docs/faq.md
@@ -32,7 +32,7 @@ When using the API, specify the `num_ctx` parameter:

 ```
 curl http://localhost:11434/api/generate -d '{
-  "model": "llama2",
+  "model": "llama3",
  "prompt": "Why is the sky blue?",
  "options": {
    "num_ctx": 4096
@@ -88,9 +88,9 @@ On windows, Ollama inherits your user and system environment variables.

 3. Edit or create New variable(s) for your user account for `OLLAMA_HOST`, `OLLAMA_MODELS`, etc.

-4. Click OK/Apply to save 
+4. Click OK/Apply to save

-5. Run `ollama` from a new terminal window 
+5. Run `ollama` from a new terminal window


 ## How can I expose Ollama on my network?
@@ -221,12 +221,12 @@ The `keep_alive` parameter can be set to:

 For example, to preload a model and leave it in memory use:
 ```shell
-curl http://localhost:11434/api/generate -d '{"model": "llama2", "keep_alive": -1}'
+curl http://localhost:11434/api/generate -d '{"model": "llama3", "keep_alive": -1}'
 ```

 To unload the model and free up memory use:
 ```shell
-curl http://localhost:11434/api/generate -d '{"model": "llama2", "keep_alive": 0}'
+curl http://localhost:11434/api/generate -d '{"model": "llama3", "keep_alive": 0}'
 ```

 Alternatively, you can change the amount of time all models are loaded into memory by setting the `OLLAMA_KEEP_ALIVE` environment variable when starting the Ollama server. The `OLLAMA_KEEP_ALIVE` variable uses the same parameter types as the `keep_alive` parameter types mentioned above. Refer to section explaining [how to configure the Ollama server](#how-do-i-configure-ollama-server) to correctly set the environment variable.
--- a/docs/modelfile.md
+++ b/docs/modelfile.md
@@ -10,7 +10,7 @@ A model file is the blueprint to create and share models with Ollama.
 - [Examples](#examples)
 - [Instructions](#instructions)
  - [FROM (Required)](#from-required)
-    - [Build from llama2](#build-from-llama2)
+    - [Build from llama3](#build-from-llama3)
    - [Build from a bin file](#build-from-a-bin-file)
  - [PARAMETER](#parameter)
    - [Valid Parameters and Values](#valid-parameters-and-values)
@@ -48,7 +48,7 @@ INSTRUCTION arguments
 An example of a `Modelfile` creating a mario blueprint:

 ```modelfile
-FROM llama2
+FROM llama3
 # sets the temperature to 1 [higher is more creative, lower is more coherent]
 PARAMETER temperature 1
 # sets the context window size to 4096, this controls how many tokens the LLM can use as context to generate the next token
@@ -67,33 +67,25 @@ To use this:

 More examples are available in the [examples directory](../examples).

-### `Modelfile`s in [ollama.com/library][1]
-
-There are two ways to view `Modelfile`s underlying the models in [ollama.com/library][1]:
-
- Option 1: view a details page from a model's tags page:
-  1.  Go to a particular model's tags (e.g. https://ollama.com/library/llama2/tags)
-  2.  Click on a tag (e.g. https://ollama.com/library/llama2:13b)
-  3.  Scroll down to "Layers"
-      - Note: if the [`FROM` instruction](#from-required) is not present,
-        it means the model was created from a local file
- Option 2: use `ollama show` to print the `Modelfile` for any local models like so:
+To view the Modelfile of a given model, use the `ollama show --modelfile` command.

  ```bash
-  > ollama show --modelfile llama2:13b
+  > ollama show --modelfile llama3
  # Modelfile generated by "ollama show"
  # To build a new Modelfile based on this one, replace the FROM line with:
-  # FROM llama2:13b
+  # FROM llama3:latest
+  FROM /Users/pdevine/.ollama/models/blobs/sha256-00e1317cbf74d901080d7100f57580ba8dd8de57203072dc6f668324ba545f29
+  TEMPLATE """{{ if .System }}<|start_header_id|>system<|end_header_id|>

-  FROM /root/.ollama/models/blobs/sha256:123abc
-  TEMPLATE """[INST] {{ if .System }}<<SYS>>{{ .System }}<</SYS>>
+  {{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|>

-  {{ end }}{{ .Prompt }} [/INST] """
-  SYSTEM """"""
-  PARAMETER stop [INST]
-  PARAMETER stop [/INST]
-  PARAMETER stop <<SYS>>
-  PARAMETER stop <</SYS>>
+  {{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|>
+
+  {{ .Response }}<|eot_id|>"""
+  PARAMETER stop "<|start_header_id|>"
+  PARAMETER stop "<|end_header_id|>"
+  PARAMETER stop "<|eot_id|>"
+  PARAMETER stop "<|reserved_special_token"
  ```

 ## Instructions
@@ -106,10 +98,10 @@ The `FROM` instruction defines the base model to use when creating a model.
 FROM <model name>:<tag>
 ```

-#### Build from llama2
+#### Build from llama3

 ```modelfile
-FROM llama2
+FROM llama3
 ```

 A list of available base models:
--- a/docs/openai.md
+++ b/docs/openai.md
@@ -25,7 +25,7 @@ chat_completion = client.chat.completions.create(
            'content': 'Say this is a test',
        }
    ],
-    model='llama2',
+    model='llama3',
 )
 ```

@@ -43,7 +43,7 @@ const openai = new OpenAI({

 const chatCompletion = await openai.chat.completions.create({
  messages: [{ role: 'user', content: 'Say this is a test' }],
-  model: 'llama2',
+  model: 'llama3',
 })
 ```

@@ -53,7 +53,7 @@ const chatCompletion = await openai.chat.completions.create({
 curl http://localhost:11434/v1/chat/completions \
    -H "Content-Type: application/json" \
    -d '{
-        "model": "llama2",
+        "model": "llama3",
        "messages": [
            {
                "role": "system",
@@ -113,7 +113,7 @@ curl http://localhost:11434/v1/chat/completions \
 Before using a model, pull it locally `ollama pull`:

 ```shell
-ollama pull llama2
+ollama pull llama3
 ```

 ### Default model names
@@ -121,7 +121,7 @@ ollama pull llama2
 For tooling that relies on default OpenAI model names such as `gpt-3.5-turbo`, use `ollama cp` to copy an existing model name to a temporary name:

 ```
-ollama cp llama2 gpt-3.5-turbo
+ollama cp llama3 gpt-3.5-turbo
 ```

 Afterwards, this new model name can be specified the `model` field:
--- a/docs/tutorials/langchainjs.md
+++ b/docs/tutorials/langchainjs.md
@@ -15,7 +15,7 @@ import { Ollama } from "langchain/llms/ollama";

 const ollama = new Ollama({
  baseUrl: "http://localhost:11434",
-  model: "llama2",
+  model: "llama3",
 });

 const answer = await ollama.invoke(`why is the sky blue?`);
@@ -23,10 +23,10 @@ const answer = await ollama.invoke(`why is the sky blue?`);
 console.log(answer);
 ```

-That will get us the same thing as if we ran `ollama run llama2 "why is the sky blue"` in the terminal. But we want to load a document from the web to ask a question against. **Cheerio** is a great library for ingesting a webpage, and **LangChain** uses it in their **CheerioWebBaseLoader**. So let's install **Cheerio** and build that part of the app.
+That will get us the same thing as if we ran `ollama run llama3 "why is the sky blue"` in the terminal. But we want to load a document from the web to ask a question against. **Cheerio** is a great library for ingesting a webpage, and **LangChain** uses it in their **CheerioWebBaseLoader**. So let's install **Cheerio** and build that part of the app.

 ```bash
-npm install cheerio 
+npm install cheerio
 ```

 ```javascript
--- a/docs/windows.md
+++ b/docs/windows.md
@@ -1,47 +1,48 @@
-# Ollama Windows Preview
-
-Welcome to the Ollama Windows preview.
-
-No more WSL required!
-
-Ollama now runs as a native Windows application, including NVIDIA and AMD Radeon GPU support.
-After installing Ollama Windows Preview, Ollama will run in the background and
-the `ollama` command line is available in `cmd`, `powershell` or your favorite
-terminal application. As usual the Ollama [api](./api.md) will be served on
-`http://localhost:11434`.
-
-As this is a preview release, you should expect a few bugs here and there.  If
-you run into a problem you can reach out on
-[Discord](https://discord.gg/ollama), or file an 
-[issue](https://github.com/ollama/ollama/issues).
-Logs will often be helpful in diagnosing the problem (see
-[Troubleshooting](#troubleshooting) below)
-
-## System Requirements
-
-* Windows 10 or newer, Home or Pro
-* NVIDIA 452.39 or newer Drivers if you have an NVIDIA card
-* AMD Radeon Driver https://www.amd.com/en/support if you have a Radeon card
-
-## API Access
-
-Here's a quick example showing API access from `powershell`
-```powershell
-(Invoke-WebRequest -method POST -Body '{"model":"llama2", "prompt":"Why is the sky blue?", "stream": false}' -uri http://localhost:11434/api/generate ).Content | ConvertFrom-json
-```
-
-## Troubleshooting
-
-While we're in preview, `OLLAMA_DEBUG` is always enabled, which adds
-a "view logs" menu item to the app, and increses logging for the GUI app and
-server.
-
-Ollama on Windows stores files in a few different locations.  You can view them in
-the explorer window by hitting `<cmd>+R` and type in:
- `explorer %LOCALAPPDATA%\Ollama` contains logs, and downloaded updates
-    - *app.log* contains logs from the GUI application
-    - *server.log* contains the server logs
-    - *upgrade.log* contains log output for upgrades
- `explorer %LOCALAPPDATA%\Programs\Ollama` contains the binaries (The installer adds this to your user PATH)
- `explorer %HOMEPATH%\.ollama` contains models and configuration
- `explorer %TEMP%` contains temporary executable files in one or more `ollama*` directories
+<<<<<<< HEAD
+# Ollama Windows Preview
+
+Welcome to the Ollama Windows preview.
+
+No more WSL required!
+
+Ollama now runs as a native Windows application, including NVIDIA and AMD Radeon GPU support.
+After installing Ollama Windows Preview, Ollama will run in the background and
+the `ollama` command line is available in `cmd`, `powershell` or your favorite
+terminal application. As usual the Ollama [api](./api.md) will be served on
+`http://localhost:11434`.
+
+As this is a preview release, you should expect a few bugs here and there.  If
+you run into a problem you can reach out on
+[Discord](https://discord.gg/ollama), or file an
+[issue](https://github.com/ollama/ollama/issues).
+Logs will often be helpful in diagnosing the problem (see
+[Troubleshooting](#troubleshooting) below)
+
+## System Requirements
+
+* Windows 10 or newer, Home or Pro
+* NVIDIA 452.39 or newer Drivers if you have an NVIDIA card
+* AMD Radeon Driver https://www.amd.com/en/support if you have a Radeon card
+
+## API Access
+
+Here's a quick example showing API access from `powershell`
+```powershell
+(Invoke-WebRequest -method POST -Body '{"model":"llama3", "prompt":"Why is the sky blue?", "stream": false}' -uri http://localhost:11434/api/generate ).Content | ConvertFrom-json
+```
+
+## Troubleshooting
+
+While we're in preview, `OLLAMA_DEBUG` is always enabled, which adds
+a "view logs" menu item to the app, and increses logging for the GUI app and
+server.
+
+Ollama on Windows stores files in a few different locations.  You can view them in
+the explorer window by hitting `<cmd>+R` and type in:
+- `explorer %LOCALAPPDATA%\Ollama` contains logs, and downloaded updates
+    - *app.log* contains logs from the GUI application
+    - *server.log* contains the server logs
+    - *upgrade.log* contains log output for upgrades
+- `explorer %LOCALAPPDATA%\Programs\Ollama` contains the binaries (The installer adds this to your user PATH)
+- `explorer %HOMEPATH%\.ollama` contains models and configuration
+- `explorer %TEMP%` contains temporary executable files in one or more `ollama*` directories
--- a/examples/bash-comparemodels/README.md
+++ b/examples/bash-comparemodels/README.md
@@ -2,7 +2,7 @@

 When calling `ollama`, you can pass it a file to run all the prompts in the file, one after the other:

-`ollama run llama2 < sourcequestions.txt`
+`ollama run llama3 < sourcequestions.txt`

 This concept is used in the following example.

--- a/examples/go-chat/main.go
+++ b/examples/go-chat/main.go
@@ -35,7 +35,7 @@ func main() {

 	ctx := context.Background()
 	req := &api.ChatRequest{
-		Model:    "llama2",
+		Model:    "llama3",
 		Messages: messages,
 	}

--- a/examples/langchain-python-rag-document/main.py
+++ b/examples/langchain-python-rag-document/main.py
@@ -40,9 +40,9 @@ while True:
        continue

    # Prompt
-    template = """Use the following pieces of context to answer the question at the end. 
-    If you don't know the answer, just say that you don't know, don't try to make up an answer. 
-    Use three sentences maximum and keep the answer as concise as possible. 
+    template = """Use the following pieces of context to answer the question at the end.
+    If you don't know the answer, just say that you don't know, don't try to make up an answer.
+    Use three sentences maximum and keep the answer as concise as possible.
    {context}
    Question: {question}
    Helpful Answer:"""
@@ -51,11 +51,11 @@ while True:
        template=template,
    )

-    llm = Ollama(model="llama2:13b", callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]))
+    llm = Ollama(model="llama3:8b", callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]))
    qa_chain = RetrievalQA.from_chain_type(
        llm,
        retriever=vectorstore.as_retriever(),
        chain_type_kwargs={"prompt": QA_CHAIN_PROMPT},
    )

-    result = qa_chain({"query": query})
+    result = qa_chain({"query": query})
--- a/examples/langchain-python-simple/README.md
+++ b/examples/langchain-python-simple/README.md
@@ -4,10 +4,10 @@ This example is a basic "hello world" of using LangChain with Ollama.

 ## Running the Example

-1. Ensure you have the `llama2` model installed:
+1. Ensure you have the `llama3` model installed:

   ```bash
-   ollama pull llama2
+   ollama pull llama3
   ```

 2. Install the Python Requirements.
@@ -21,4 +21,3 @@ This example is a basic "hello world" of using LangChain with Ollama.
   ```bash
   python main.py
   ```
-  
--- a/examples/langchain-python-simple/main.py
+++ b/examples/langchain-python-simple/main.py
@@ -1,6 +1,6 @@
 from langchain.llms import Ollama

 input = input("What is your question?")
-llm = Ollama(model="llama2")
+llm = Ollama(model="llama3")
 res = llm.predict(input)
 print (res)
--- a/examples/modelfile-mario/Modelfile
+++ b/examples/modelfile-mario/Modelfile
@@ -1,4 +1,4 @@
-FROM llama2
+FROM llama3
 PARAMETER temperature 1
 SYSTEM """
 You are Mario from super mario bros, acting as an assistant.
--- a/examples/modelfile-mario/readme.md
+++ b/examples/modelfile-mario/readme.md
@@ -2,12 +2,12 @@

 # Example character: Mario

-This example shows how to create a basic character using Llama2 as the base model.
+This example shows how to create a basic character using Llama3 as the base model.

 To run this example:

 1. Download the Modelfile
-2. `ollama pull llama2` to get the base model used in the model file.
+2. `ollama pull llama3` to get the base model used in the model file.
 3. `ollama create NAME -f ./Modelfile`
 4. `ollama run NAME`

@@ -18,7 +18,7 @@ Ask it some questions like "Who are you?" or "Is Peach in trouble again?"
 What the model file looks like:

 ```
-FROM llama2
+FROM llama3
 PARAMETER temperature 1
 SYSTEM """
 You are Mario from Super Mario Bros, acting as an assistant.
--- a/examples/python-json-datagenerator/predefinedschema.py
+++ b/examples/python-json-datagenerator/predefinedschema.py
@@ -2,16 +2,16 @@ import requests
 import json
 import random

-model = "llama2"
+model = "llama3"
 template = {
-  "firstName": "", 
-  "lastName": "", 
+  "firstName": "",
+  "lastName": "",
  "address": {
-    "street": "", 
-    "city": "", 
-    "state": "", 
+    "street": "",
+    "city": "",
+    "state": "",
    "zipCode": ""
-  }, 
+  },
  "phoneNumber": ""
 }

--- a/examples/python-json-datagenerator/randomaddresses.py
+++ b/examples/python-json-datagenerator/randomaddresses.py
@@ -12,7 +12,7 @@ countries = [
    "France",
 ]
 country = random.choice(countries)
-model = "llama2"
+model = "llama3"

 prompt = f"generate one realistically believable sample data set of a persons first name, last name, address in {country}, and phone number. Do not use common names. Respond using JSON. Key names should have no backslashes, values should use plain ascii with no special characters."

--- a/examples/python-json-datagenerator/readme.md
+++ b/examples/python-json-datagenerator/readme.md
@@ -6,10 +6,10 @@ There are two python scripts in this example. `randomaddresses.py` generates ran

 ## Running the Example

-1. Ensure you have the `llama2` model installed:
+1. Ensure you have the `llama3` model installed:

   ```bash
-   ollama pull llama2
+   ollama pull llama3
   ```

 2. Install the Python Requirements.
--- a/examples/python-simplechat/client.py
+++ b/examples/python-simplechat/client.py
@@ -2,7 +2,7 @@ import json
 import requests

 # NOTE: ollama must be running for this to work, start the ollama app or run `ollama serve`
-model = "llama2"  # TODO: update this for whatever model you wish to use
+model = "llama3"  # TODO: update this for whatever model you wish to use


 def chat(messages):
--- a/examples/python-simplechat/readme.md
+++ b/examples/python-simplechat/readme.md
@@ -4,10 +4,10 @@ The **chat** endpoint is one of two ways to generate text from an LLM with Ollam

 ## Running the Example

-1. Ensure you have the `llama2` model installed:
+1. Ensure you have the `llama3` model installed:

   ```bash
-   ollama pull llama2
+   ollama pull llama3
   ```

 2. Install the Python Requirements.
--- a/examples/typescript-mentors/README.md
+++ b/examples/typescript-mentors/README.md
@@ -4,10 +4,10 @@ This example demonstrates how one would create a set of 'mentors' you can have a

 ## Usage

-1. Add llama2 to have the mentors ask your questions:
+1. Add llama3 to have the mentors ask your questions:

   ```bash
-   ollama pull llama2
+   ollama pull llama3
   ```

 2. Install prerequisites:
--- a/examples/typescript-mentors/character-generator.ts
+++ b/examples/typescript-mentors/character-generator.ts
@@ -15,7 +15,7 @@ async function characterGenerator() {
  ollama.setModel("stablebeluga2:70b-q4_K_M");
  const bio = await ollama.generate(`create a bio of ${character} in a single long paragraph. Instead of saying '${character} is...' or '${character} was...' use language like 'You are...' or 'You were...'. Then create a paragraph describing the speaking mannerisms and style of ${character}. Don't include anything about how ${character} looked or what they sounded like, just focus on the words they said. Instead of saying '${character} would say...' use language like 'You should say...'. If you use quotes, always use single quotes instead of double quotes. If there are any specific words or phrases you used a lot, show how you used them. `);

-  const thecontents = `FROM llama2\nSYSTEM """\n${bio.response.replace(/(\r\n|\n|\r)/gm, " ").replace('would', 'should')} All answers to questions should be related back to what you are most known for.\n"""`;
+  const thecontents = `FROM llama3\nSYSTEM """\n${bio.response.replace(/(\r\n|\n|\r)/gm, " ").replace('would', 'should')} All answers to questions should be related back to what you are most known for.\n"""`;

  fs.writeFile(path.join(directory, 'Modelfile'), thecontents, (err: any) => {
    if (err) throw err;
@@ -23,4 +23,4 @@ async function characterGenerator() {
  });
 }

-characterGenerator();
+characterGenerator();
--- a/examples/typescript-simplechat/client.ts
+++ b/examples/typescript-simplechat/client.ts
@@ -1,6 +1,6 @@
 import * as readline from "readline";

-const model = "llama2";
+const model = "llama3";
 type Message = {
  role: "assistant" | "user" | "system";
  content: string;
@@ -74,4 +74,4 @@ async function main() {

 }

-main();
+main();
--- a/integration/utils_test.go
+++ b/integration/utils_test.go
@@ -107,7 +107,7 @@ func startServer(ctx context.Context, ollamaHost string) error {

 	if tmp := os.Getenv("OLLAMA_HOST"); tmp != ollamaHost {
 		slog.Info("setting env", "OLLAMA_HOST", ollamaHost)
-		os.Setenv("OLLAMA_HOST", ollamaHost)
+		t.Setenv("OLLAMA_HOST", ollamaHost)
 	}

 	slog.Info("starting server", "url", ollamaHost)
--- a/llm/filetype.go
+++ b/llm/filetype.go
@@ -0,0 +1,165 @@
+package llm
+
+import "fmt"
+
+type fileType uint32
+
+const (
+	fileTypeF32 fileType = iota
+	fileTypeF16
+	fileTypeQ4_0
+	fileTypeQ4_1
+	fileTypeQ4_1_F16
+	fileTypeQ4_2 // unused
+	fileTypeQ4_3 // unused
+	fileTypeQ8_0
+	fileTypeQ5_0
+	fileTypeQ5_1
+	fileTypeQ2_K
+	fileTypeQ3_K_S
+	fileTypeQ3_K_M
+	fileTypeQ3_K_L
+	fileTypeQ4_K_S
+	fileTypeQ4_K_M
+	fileTypeQ5_K_S
+	fileTypeQ5_K_M
+	fileTypeQ6_K
+	fileTypeIQ2_XXS
+	fileTypeIQ2_XS
+	fileTypeQ2_K_S
+	fileTypeQ3_K_XS
+	fileTypeIQ3_XXS
+	fileTypeIQ1_S
+	fileTypeIQ4_NL
+	fileTypeIQ3_S
+	fileTypeIQ2_S
+	fileTypeIQ4_XS
+
+	fileTypeUnknown
+)
+
+func ParseFileType(s string) (fileType, error) {
+	switch s {
+	case "F32":
+		return fileTypeF32, nil
+	case "F16":
+		return fileTypeF16, nil
+	case "Q4_0":
+		return fileTypeQ4_0, nil
+	case "Q4_1":
+		return fileTypeQ4_1, nil
+	case "Q4_1_F16":
+		return fileTypeQ4_1_F16, nil
+	case "Q8_0":
+		return fileTypeQ8_0, nil
+	case "Q5_0":
+		return fileTypeQ5_0, nil
+	case "Q5_1":
+		return fileTypeQ5_1, nil
+	case "Q2_K":
+		return fileTypeQ2_K, nil
+	case "Q3_K_S":
+		return fileTypeQ3_K_S, nil
+	case "Q3_K_M":
+		return fileTypeQ3_K_M, nil
+	case "Q3_K_L":
+		return fileTypeQ3_K_L, nil
+	case "Q4_K_S":
+		return fileTypeQ4_K_S, nil
+	case "Q4_K_M":
+		return fileTypeQ4_K_M, nil
+	case "Q5_K_S":
+		return fileTypeQ5_K_S, nil
+	case "Q5_K_M":
+		return fileTypeQ5_K_M, nil
+	case "Q6_K":
+		return fileTypeQ6_K, nil
+	case "IQ2_XXS":
+		return fileTypeIQ2_XXS, nil
+	case "IQ2_XS":
+		return fileTypeIQ2_XS, nil
+	case "Q2_K_S":
+		return fileTypeQ2_K_S, nil
+	case "Q3_K_XS":
+		return fileTypeQ3_K_XS, nil
+	case "IQ1_S":
+		return fileTypeIQ1_S, nil
+	case "IQ4_NL":
+		return fileTypeIQ4_NL, nil
+	case "IQ3_S":
+		return fileTypeIQ3_S, nil
+	case "IQ2_S":
+		return fileTypeIQ2_S, nil
+	case "IQ4_XS":
+		return fileTypeIQ4_XS, nil
+	case "IQ3_XXS":
+		return fileTypeIQ3_XXS, nil
+	default:
+		return fileTypeUnknown, fmt.Errorf("unknown fileType: %s", s)
+	}
+}
+
+func (t fileType) String() string {
+	switch t {
+	case fileTypeF32:
+		return "F32"
+	case fileTypeF16:
+		return "F16"
+	case fileTypeQ4_0:
+		return "Q4_0"
+	case fileTypeQ4_1:
+		return "Q4_1"
+	case fileTypeQ4_1_F16:
+		return "Q4_1_F16"
+	case fileTypeQ8_0:
+		return "Q8_0"
+	case fileTypeQ5_0:
+		return "Q5_0"
+	case fileTypeQ5_1:
+		return "Q5_1"
+	case fileTypeQ2_K:
+		return "Q2_K"
+	case fileTypeQ3_K_S:
+		return "Q3_K_S"
+	case fileTypeQ3_K_M:
+		return "Q3_K_M"
+	case fileTypeQ3_K_L:
+		return "Q3_K_L"
+	case fileTypeQ4_K_S:
+		return "Q4_K_S"
+	case fileTypeQ4_K_M:
+		return "Q4_K_M"
+	case fileTypeQ5_K_S:
+		return "Q5_K_S"
+	case fileTypeQ5_K_M:
+		return "Q5_K_M"
+	case fileTypeQ6_K:
+		return "Q6_K"
+	case fileTypeIQ2_XXS:
+		return "IQ2_XXS"
+	case fileTypeIQ2_XS:
+		return "IQ2_XS"
+	case fileTypeQ2_K_S:
+		return "Q2_K_S"
+	case fileTypeQ3_K_XS:
+		return "Q3_K_XS"
+	case fileTypeIQ1_S:
+		return "IQ1_S"
+	case fileTypeIQ4_NL:
+		return "IQ4_NL"
+	case fileTypeIQ3_S:
+		return "IQ3_S"
+	case fileTypeIQ2_S:
+		return "IQ2_S"
+	case fileTypeIQ4_XS:
+		return "IQ4_XS"
+	case fileTypeIQ3_XXS:
+		return "IQ3_XXS"
+	default:
+		return "unknown"
+	}
+}
+
+func (t fileType) Value() uint32 {
+	return uint32(t)
+}
--- a/llm/ggml.go
+++ b/llm/ggml.go
@@ -13,82 +13,6 @@ type GGML struct {
 	model
 }

-const (
-	fileTypeF32 uint32 = iota
-	fileTypeF16
-	fileTypeQ4_0
-	fileTypeQ4_1
-	fileTypeQ4_1_F16
-	fileTypeQ8_0 uint32 = iota + 2
-	fileTypeQ5_0
-	fileTypeQ5_1
-	fileTypeQ2_K
-	fileTypeQ3_K_S
-	fileTypeQ3_K_M
-	fileTypeQ3_K_L
-	fileTypeQ4_K_S
-	fileTypeQ4_K_M
-	fileTypeQ5_K_S
-	fileTypeQ5_K_M
-	fileTypeQ6_K
-	fileTypeIQ2_XXS
-	fileTypeIQ2_XS
-	fileTypeQ2_K_S
-	fileTypeQ3_K_XS
-	fileTypeIQ3_XXS
-)
-
-func fileType(fileType uint32) string {
-	switch fileType {
-	case fileTypeF32:
-		return "F32"
-	case fileTypeF16:
-		return "F16"
-	case fileTypeQ4_0:
-		return "Q4_0"
-	case fileTypeQ4_1:
-		return "Q4_1"
-	case fileTypeQ4_1_F16:
-		return "Q4_1_F16"
-	case fileTypeQ8_0:
-		return "Q8_0"
-	case fileTypeQ5_0:
-		return "Q5_0"
-	case fileTypeQ5_1:
-		return "Q5_1"
-	case fileTypeQ2_K:
-		return "Q2_K"
-	case fileTypeQ3_K_S:
-		return "Q3_K_S"
-	case fileTypeQ3_K_M:
-		return "Q3_K_M"
-	case fileTypeQ3_K_L:
-		return "Q3_K_L"
-	case fileTypeQ4_K_S:
-		return "Q4_K_S"
-	case fileTypeQ4_K_M:
-		return "Q4_K_M"
-	case fileTypeQ5_K_S:
-		return "Q5_K_S"
-	case fileTypeQ5_K_M:
-		return "Q5_K_M"
-	case fileTypeQ6_K:
-		return "Q6_K"
-	case fileTypeIQ2_XXS:
-		return "IQ2_XXS"
-	case fileTypeIQ2_XS:
-		return "IQ2_XS"
-	case fileTypeQ2_K_S:
-		return "Q2_K_S"
-	case fileTypeQ3_K_XS:
-		return "Q3_K_XS"
-	case fileTypeIQ3_XXS:
-		return "IQ3_XXS"
-	default:
-		return "unknown"
-	}
-}
-
 type model interface {
 	KV() KV
 	Tensors() Tensors
@@ -123,7 +47,7 @@ func (kv KV) ParameterCount() uint64 {

 func (kv KV) FileType() string {
 	if u64 := kv.u64("general.file_type"); u64 > 0 {
-		return fileType(uint32(u64))
+		return fileType(uint32(u64)).String()
 	}

 	return "unknown"
@@ -201,9 +125,9 @@ type Tensor struct {

 func (t Tensor) blockSize() uint64 {
 	switch {
-	case t.Kind < 2:
+	case t.Kind < 2 || (t.Kind > 23 && t.Kind < 29):
 		return 1
-	case t.Kind < 10:
+	case t.Kind < 10 || t.Kind == 20:
 		return 32
 	default:
 		return 256
@@ -248,6 +172,16 @@ func (t Tensor) typeSize() uint64 {
 		return 2 + 2*blockSize/8 + blockSize/32
 	case 18: // IQ3_XXS
 		return 2 + 3*blockSize/8
+	case 19: // IQ1_S
+		return 2 + blockSize/8 + blockSize/16
+	case 20: // IQ4_NL
+		return 2 + blockSize/2
+	case 21: // IQ3_S
+		return 2 + 2*blockSize/8 + blockSize/8 + blockSize/32 + 4
+	case 22: // IQ2_S
+		return 2 + blockSize/4 + blockSize/16
+	case 23: // IQ4_XS
+		return 4 + blockSize/2 + blockSize/64
 	default:
 		return 0
 	}
@@ -286,6 +220,23 @@ const (

 var ErrUnsupportedFormat = errors.New("unsupported model format")

+func DetectGGMLType(b []byte) string {
+	switch binary.LittleEndian.Uint32(b[:4]) {
+	case FILE_MAGIC_GGML:
+		return "ggml"
+	case FILE_MAGIC_GGMF:
+		return "ggmf"
+	case FILE_MAGIC_GGJT:
+		return "ggjt"
+	case FILE_MAGIC_GGLA:
+		return "ggla"
+	case FILE_MAGIC_GGUF_LE, FILE_MAGIC_GGUF_BE:
+		return "gguf"
+	default:
+		return ""
+	}
+}
+
 func DecodeGGML(rs io.ReadSeeker) (*GGML, int64, error) {
 	var magic uint32
 	if err := binary.Read(rs, binary.LittleEndian, &magic); err != nil {
--- a/llm/llm.go
+++ b/llm/llm.go
@@ -20,7 +20,7 @@ func SystemInfo() string {
 	return C.GoString(C.llama_print_system_info())
 }

-func Quantize(infile, outfile, filetype string) error {
+func Quantize(infile, outfile string, ftype fileType) error {
 	cinfile := C.CString(infile)
 	defer C.free(unsafe.Pointer(cinfile))

@@ -29,58 +29,10 @@ func Quantize(infile, outfile, filetype string) error {

 	params := C.llama_model_quantize_default_params()
 	params.nthread = -1
+	params.ftype = ftype.Value()

-	switch filetype {
-	case "F32":
-		params.ftype = fileTypeF32
-	case "F16":
-		params.ftype = fileTypeF16
-	case "Q4_0":
-		params.ftype = fileTypeQ4_0
-	case "Q4_1":
-		params.ftype = fileTypeQ4_1
-	case "Q4_1_F16":
-		params.ftype = fileTypeQ4_1_F16
-	case "Q8_0":
-		params.ftype = fileTypeQ8_0
-	case "Q5_0":
-		params.ftype = fileTypeQ5_0
-	case "Q5_1":
-		params.ftype = fileTypeQ5_1
-	case "Q2_K":
-		params.ftype = fileTypeQ2_K
-	case "Q3_K_S":
-		params.ftype = fileTypeQ3_K_S
-	case "Q3_K_M":
-		params.ftype = fileTypeQ3_K_M
-	case "Q3_K_L":
-		params.ftype = fileTypeQ3_K_L
-	case "Q4_K_S":
-		params.ftype = fileTypeQ4_K_S
-	case "Q4_K_M":
-		params.ftype = fileTypeQ4_K_M
-	case "Q5_K_S":
-		params.ftype = fileTypeQ5_K_S
-	case "Q5_K_M":
-		params.ftype = fileTypeQ5_K_M
-	case "Q6_K":
-		params.ftype = fileTypeQ6_K
-	case "IQ2_XXS":
-		params.ftype = fileTypeIQ2_XXS
-	case "IQ2_XS":
-		params.ftype = fileTypeIQ2_XS
-	case "Q2_K_S":
-		params.ftype = fileTypeQ2_K_S
-	case "Q3_K_XS":
-		params.ftype = fileTypeQ3_K_XS
-	case "IQ3_XXS":
-		params.ftype = fileTypeIQ3_XXS
-	default:
-		return fmt.Errorf("unknown filetype: %s", filetype)
-	}
-
-	if retval := C.llama_model_quantize(cinfile, coutfile, &params); retval != 0 {
-		return fmt.Errorf("llama_model_quantize: %d", retval)
+	if rc := C.llama_model_quantize(cinfile, coutfile, &params); rc != 0 {
+		return fmt.Errorf("llama_model_quantize: %d", rc)
 	}

 	return nil
--- a/parser/parser.go
+++ b/parser/parser.go
@@ -1,132 +0,0 @@
-package parser
-
-import (
-	"bufio"
-	"bytes"
-	"errors"
-	"fmt"
-	"io"
-	"log/slog"
-	"slices"
-)
-
-type Command struct {
-	Name string
-	Args string
-}
-
-func (c *Command) Reset() {
-	c.Name = ""
-	c.Args = ""
-}
-
-func Parse(reader io.Reader) ([]Command, error) {
-	var commands []Command
-	var command, modelCommand Command
-
-	scanner := bufio.NewScanner(reader)
-	scanner.Buffer(make([]byte, 0, bufio.MaxScanTokenSize), bufio.MaxScanTokenSize)
-	scanner.Split(scanModelfile)
-	for scanner.Scan() {
-		line := scanner.Bytes()
-
-		fields := bytes.SplitN(line, []byte(" "), 2)
-		if len(fields) == 0 || len(fields[0]) == 0 {
-			continue
-		}
-
-		switch string(bytes.ToUpper(fields[0])) {
-		case "FROM":
-			command.Name = "model"
-			command.Args = string(bytes.TrimSpace(fields[1]))
-			// copy command for validation
-			modelCommand = command
-		case "ADAPTER":
-			command.Name = string(bytes.ToLower(fields[0]))
-			command.Args = string(bytes.TrimSpace(fields[1]))
-		case "LICENSE", "TEMPLATE", "SYSTEM", "PROMPT":
-			command.Name = string(bytes.ToLower(fields[0]))
-			command.Args = string(fields[1])
-		case "PARAMETER":
-			fields = bytes.SplitN(fields[1], []byte(" "), 2)
-			if len(fields) < 2 {
-				return nil, fmt.Errorf("missing value for %s", fields)
-			}
-
-			command.Name = string(fields[0])
-			command.Args = string(bytes.TrimSpace(fields[1]))
-		case "EMBED":
-			return nil, fmt.Errorf("deprecated command: EMBED is no longer supported, use the /embed API endpoint instead")
-		case "MESSAGE":
-			command.Name = string(bytes.ToLower(fields[0]))
-			fields = bytes.SplitN(fields[1], []byte(" "), 2)
-			if len(fields) < 2 {
-				return nil, fmt.Errorf("should be in the format <role> <message>")
-			}
-			if !slices.Contains([]string{"system", "user", "assistant"}, string(bytes.ToLower(fields[0]))) {
-				return nil, fmt.Errorf("role must be one of \"system\", \"user\", or \"assistant\"")
-			}
-			command.Args = fmt.Sprintf("%s: %s", string(bytes.ToLower(fields[0])), string(fields[1]))
-		default:
-			if !bytes.HasPrefix(fields[0], []byte("#")) {
-				// log a warning for unknown commands
-				slog.Warn(fmt.Sprintf("Unknown command: %s", fields[0]))
-			}
-			continue
-		}
-
-		commands = append(commands, command)
-		command.Reset()
-	}
-
-	if modelCommand.Args == "" {
-		return nil, errors.New("no FROM line for the model was specified")
-	}
-
-	return commands, scanner.Err()
-}
-
-func scanModelfile(data []byte, atEOF bool) (advance int, token []byte, err error) {
-	advance, token, err = scan([]byte(`"""`), []byte(`"""`), data, atEOF)
-	if err != nil {
-		return 0, nil, err
-	}
-
-	if advance > 0 && token != nil {
-		return advance, token, nil
-	}
-
-	advance, token, err = scan([]byte(`"`), []byte(`"`), data, atEOF)
-	if err != nil {
-		return 0, nil, err
-	}
-
-	if advance > 0 && token != nil {
-		return advance, token, nil
-	}
-
-	return bufio.ScanLines(data, atEOF)
-}
-
-func scan(openBytes, closeBytes, data []byte, atEOF bool) (advance int, token []byte, err error) {
-	newline := bytes.IndexByte(data, '\n')
-
-	if start := bytes.Index(data, openBytes); start >= 0 && start < newline {
-		end := bytes.Index(data[start+len(openBytes):], closeBytes)
-		if end < 0 {
-			if atEOF {
-				return 0, nil, fmt.Errorf("unterminated %s: expecting %s", openBytes, closeBytes)
-			} else {
-				return 0, nil, nil
-			}
-		}
-
-		n := start + len(openBytes) + end + len(closeBytes)
-
-		newData := data[:start]
-		newData = append(newData, data[start+len(openBytes):n-len(closeBytes)]...)
-		return n, newData, nil
-	}
-
-	return 0, nil, nil
-}
--- a/parser/parser_test.go
+++ b/parser/parser_test.go
@@ -1,98 +0,0 @@
-package parser
-
-import (
-	"strings"
-	"testing"
-
-	"github.com/stretchr/testify/assert"
-)
-
-func Test_Parser(t *testing.T) {
-
-	input := `
-FROM model1
-ADAPTER adapter1
-LICENSE MIT
-PARAMETER param1 value1
-PARAMETER param2 value2
-TEMPLATE template1
-`
-
-	reader := strings.NewReader(input)
-
-	commands, err := Parse(reader)
-	assert.Nil(t, err)
-
-	expectedCommands := []Command{
-		{Name: "model", Args: "model1"},
-		{Name: "adapter", Args: "adapter1"},
-		{Name: "license", Args: "MIT"},
-		{Name: "param1", Args: "value1"},
-		{Name: "param2", Args: "value2"},
-		{Name: "template", Args: "template1"},
-	}
-
-	assert.Equal(t, expectedCommands, commands)
-}
-
-func Test_Parser_NoFromLine(t *testing.T) {
-
-	input := `
-PARAMETER param1 value1
-PARAMETER param2 value2
-`
-
-	reader := strings.NewReader(input)
-
-	_, err := Parse(reader)
-	assert.ErrorContains(t, err, "no FROM line")
-}
-
-func Test_Parser_MissingValue(t *testing.T) {
-
-	input := `
-FROM foo
-PARAMETER param1
-`
-
-	reader := strings.NewReader(input)
-
-	_, err := Parse(reader)
-	assert.ErrorContains(t, err, "missing value for [param1]")
-
-}
-
-func Test_Parser_Messages(t *testing.T) {
-
-	input := `
-FROM foo
-MESSAGE system You are a Parser. Always Parse things.
-MESSAGE user Hey there!
-MESSAGE assistant Hello, I want to parse all the things!
-`
-
-	reader := strings.NewReader(input)
-	commands, err := Parse(reader)
-	assert.Nil(t, err)
-
-	expectedCommands := []Command{
-		{Name: "model", Args: "foo"},
-		{Name: "message", Args: "system: You are a Parser. Always Parse things."},
-		{Name: "message", Args: "user: Hey there!"},
-		{Name: "message", Args: "assistant: Hello, I want to parse all the things!"},
-	}
-
-	assert.Equal(t, expectedCommands, commands)
-}
-
-func Test_Parser_Messages_BadRole(t *testing.T) {
-
-	input := `
-FROM foo
-MESSAGE badguy I'm a bad guy!
-`
-
-	reader := strings.NewReader(input)
-	_, err := Parse(reader)
-	assert.ErrorContains(t, err, "role must be one of \"system\", \"user\", or \"assistant\"")
-}
--- a/server/images.go
+++ b/server/images.go
@@ -1,8 +1,8 @@
 package server

 import (
-	"archive/zip"
 	"bytes"
+	"cmp"
 	"context"
 	"crypto/sha256"
 	"encoding/base64"
@@ -11,7 +11,6 @@ import (
 	"errors"
 	"fmt"
 	"io"
-	"io/fs"
 	"log"
 	"log/slog"
 	"net/http"
@@ -21,16 +20,13 @@ import (
 	"runtime"
 	"strconv"
 	"strings"
-	"text/template"

 	"golang.org/x/exp/slices"

 	"github.com/ollama/ollama/api"
 	"github.com/ollama/ollama/auth"
-	"github.com/ollama/ollama/convert"
 	"github.com/ollama/ollama/format"
 	"github.com/ollama/ollama/llm"
-	"github.com/ollama/ollama/parser"
 	"github.com/ollama/ollama/types/errtypes"
 	"github.com/ollama/ollama/types/model"
 	"github.com/ollama/ollama/version"
@@ -64,6 +60,76 @@ func (m *Model) IsEmbedding() bool {
 	return slices.Contains(m.Config.ModelFamilies, "bert") || slices.Contains(m.Config.ModelFamilies, "nomic-bert")
 }

+func (m *Model) String() string {
+	var modelfile model.File
+
+	modelfile.Commands = append(modelfile.Commands, model.Command{
+		Name: "model",
+		Args: m.ModelPath,
+	})
+
+	if m.Template != "" {
+		modelfile.Commands = append(modelfile.Commands, model.Command{
+			Name: "template",
+			Args: m.Template,
+		})
+	}
+
+	if m.System != "" {
+		modelfile.Commands = append(modelfile.Commands, model.Command{
+			Name: "system",
+			Args: m.System,
+		})
+	}
+
+	for _, adapter := range m.AdapterPaths {
+		modelfile.Commands = append(modelfile.Commands, model.Command{
+			Name: "adapter",
+			Args: adapter,
+		})
+	}
+
+	for _, projector := range m.ProjectorPaths {
+		modelfile.Commands = append(modelfile.Commands, model.Command{
+			Name: "projector",
+			Args: projector,
+		})
+	}
+
+	for k, v := range m.Options {
+		switch v := v.(type) {
+		case []any:
+			for _, s := range v {
+				modelfile.Commands = append(modelfile.Commands, model.Command{
+					Name: k,
+					Args: fmt.Sprintf("%v", s),
+				})
+			}
+		default:
+			modelfile.Commands = append(modelfile.Commands, model.Command{
+				Name: k,
+				Args: fmt.Sprintf("%v", v),
+			})
+		}
+	}
+
+	for _, license := range m.License {
+		modelfile.Commands = append(modelfile.Commands, model.Command{
+			Name: "license",
+			Args: license,
+		})
+	}
+
+	for _, msg := range m.Messages {
+		modelfile.Commands = append(modelfile.Commands, model.Command{
+			Name: "message",
+			Args: fmt.Sprintf("%s %s", msg.Role, msg.Content),
+		})
+	}
+
+	return modelfile.String()
+}
+
 type Message struct {
 	Role    string `json:"role"`
 	Content string `json:"content"`
@@ -89,36 +155,6 @@ type ConfigV2 struct {
 	RootFS       RootFS `json:"rootfs"`
 }

-func (c *ConfigV2) SetModelFormat(format string) {
-	if c.ModelFormat == "" {
-		c.ModelFormat = format
-	}
-}
-
-func (c *ConfigV2) SetModelFamily(families ...string) {
-	for _, family := range families {
-		if c.ModelFamily == "" {
-			c.ModelFamily = family
-		}
-
-		if !slices.Contains(c.ModelFamilies, family) {
-			c.ModelFamilies = append(c.ModelFamilies, family)
-		}
-	}
-}
-
-func (c *ConfigV2) SetModelType(modelType string) {
-	if c.ModelType == "" {
-		c.ModelType = modelType
-	}
-}
-
-func (c *ConfigV2) SetFileType(fileType string) {
-	if c.FileType == "" {
-		c.FileType = fileType
-	}
-}
-
 type RootFS struct {
 	Type    string   `json:"type"`
 	DiffIDs []string `json:"diff_ids"`
@@ -263,7 +299,7 @@ func GetModel(name string) (*Model, error) {
 	return model, nil
 }

-func realpath(mfDir, from string) string {
+func realpath(rel, from string) string {
 	abspath, err := filepath.Abs(from)
 	if err != nil {
 		return from
@@ -280,22 +316,15 @@ func realpath(mfDir, from string) string {
 		return filepath.Join(home, from[2:])
 	}

-	if _, err := os.Stat(filepath.Join(mfDir, from)); err == nil {
+	if _, err := os.Stat(filepath.Join(rel, from)); err == nil {
 		// this is a file relative to the Modelfile
-		return filepath.Join(mfDir, from)
+		return filepath.Join(rel, from)
 	}

 	return abspath
 }

-func CreateModel(ctx context.Context, name, modelFileDir, quantization string, commands []parser.Command, fn func(resp api.ProgressResponse)) error {
-	deleteMap := make(map[string]struct{})
-	if manifest, _, err := GetManifest(ParseModelPath(name)); err == nil {
-		for _, layer := range append(manifest.Layers, manifest.Config) {
-			deleteMap[layer.Digest] = struct{}{}
-		}
-	}
-
+func CreateModel(ctx context.Context, name, modelFileDir, quantization string, modelfile *model.File, fn func(resp api.ProgressResponse)) (err error) {
 	config := ConfigV2{
 		OS:           "linux",
 		Architecture: "amd64",
@@ -304,250 +333,181 @@ func CreateModel(ctx context.Context, name, modelFileDir, quantization string, c
 		},
 	}

-	var layers Layers
-	messages := []string{}
+	var messages []*api.Message
+	parameters := make(map[string]any)

-	params := make(map[string][]string)
-	fromParams := make(map[string]any)
-
-	for _, c := range commands {
+	var layers []*Layer
+	for _, c := range modelfile.Commands {
 		mediatype := fmt.Sprintf("application/vnd.ollama.image.%s", c.Name)

 		switch c.Name {
-		case "model":
-			if strings.HasPrefix(c.Args, "@") {
-				blobPath, err := GetBlobsPath(strings.TrimPrefix(c.Args, "@"))
+		case "model", "adapter":
+			var baseLayers []*layerWithGGML
+			if name := model.ParseName(c.Args); name.IsValid() {
+				baseLayers, err = parseFromModel(ctx, name, fn)
+				if err != nil {
+					return err
+				}
+			} else if strings.HasPrefix(c.Args, "@") {
+				blobpath, err := GetBlobsPath(strings.TrimPrefix(c.Args, "@"))
 				if err != nil {
 					return err
 				}

-				c.Args = blobPath
-			}
-
-			pathName := realpath(modelFileDir, c.Args)
-
-			ggufName, err := convertModel(name, pathName, fn)
-			if err != nil {
-				var pathErr *fs.PathError
-				switch {
-				case errors.Is(err, zip.ErrFormat):
-					// it's not a safetensor archive
-				case errors.As(err, &pathErr):
-					// it's not a file on disk, could be a model reference
-				default:
-					return err
-				}
-			}
-
-			if ggufName != "" {
-				pathName = ggufName
-				defer os.RemoveAll(ggufName)
-
-				if quantization != "" {
-					quantization = strings.ToUpper(quantization)
-					fn(api.ProgressResponse{Status: fmt.Sprintf("quantizing %s model to %s", "F16", quantization)})
-					tempfile, err := os.CreateTemp(filepath.Dir(ggufName), quantization)
-					if err != nil {
-						return err
-					}
-					defer os.RemoveAll(tempfile.Name())
-
-					if err := llm.Quantize(ggufName, tempfile.Name(), quantization); err != nil {
-						return err
-					}
-
-					if err := tempfile.Close(); err != nil {
-						return err
-					}
-
-					pathName = tempfile.Name()
-				}
-			}
-
-			bin, err := os.Open(pathName)
-			if err != nil {
-				// not a file on disk so must be a model reference
-				modelpath := ParseModelPath(c.Args)
-				manifest, _, err := GetManifest(modelpath)
-				switch {
-				case errors.Is(err, os.ErrNotExist):
-					fn(api.ProgressResponse{Status: "pulling model"})
-					if err := PullModel(ctx, c.Args, &registryOptions{}, fn); err != nil {
-						return err
-					}
-
-					manifest, _, err = GetManifest(modelpath)
-					if err != nil {
-						return err
-					}
-				case err != nil:
-					return err
-				}
-
-				fn(api.ProgressResponse{Status: "reading model metadata"})
-				fromConfigPath, err := GetBlobsPath(manifest.Config.Digest)
+				blob, err := os.Open(blobpath)
 				if err != nil {
 					return err
 				}
+				defer blob.Close()

-				fromConfigFile, err := os.Open(fromConfigPath)
+				baseLayers, err = parseFromFile(ctx, blob, fn)
 				if err != nil {
 					return err
 				}
-				defer fromConfigFile.Close()
+			} else if file, err := os.Open(realpath(modelFileDir, c.Args)); err == nil {
+				defer file.Close()

-				var fromConfig ConfigV2
-				if err := json.NewDecoder(fromConfigFile).Decode(&fromConfig); err != nil {
+				baseLayers, err = parseFromFile(ctx, file, fn)
+				if err != nil {
 					return err
 				}
+			} else {
+				return fmt.Errorf("invalid model reference: %s", c.Args)
+			}

-				// if the model is still not in gguf format, error out
-				if fromConfig.ModelFormat != "gguf" {
-					return fmt.Errorf("%s is not in gguf format, this base model is not compatible with this version of ollama", c.Args)
-				}
-
-				config.SetModelFormat(fromConfig.ModelFormat)
-				config.SetModelFamily(append(fromConfig.ModelFamilies, fromConfig.ModelFamily)...)
-				config.SetModelType(fromConfig.ModelType)
-				config.SetFileType(fromConfig.FileType)
-
-				for _, layer := range manifest.Layers {
-					deleteMap[layer.Digest] = struct{}{}
-					if layer.MediaType == "application/vnd.ollama.image.params" {
-						fromParamsPath, err := GetBlobsPath(layer.Digest)
-						if err != nil {
-							return err
-						}
-
-						fromParamsFile, err := os.Open(fromParamsPath)
-						if err != nil {
-							return err
-						}
-						defer fromParamsFile.Close()
-
-						if err := json.NewDecoder(fromParamsFile).Decode(&fromParams); err != nil {
-							return err
-						}
-					}
-
-					layer, err := NewLayerFromLayer(layer.Digest, layer.MediaType, modelpath.GetShortTagname())
+			for _, baseLayer := range baseLayers {
+				if quantization != "" &&
+					baseLayer.MediaType == "application/vnd.ollama.image.model" &&
+					baseLayer.GGML != nil &&
+					baseLayer.GGML.Name() == "gguf" {
+					ftype, err := llm.ParseFileType(quantization)
 					if err != nil {
 						return err
 					}

-					layers.Add(layer)
+					filetype := baseLayer.GGML.KV().FileType()
+					if !slices.Contains([]string{"F16", "F32"}, filetype) {
+						return errors.New("quantization is only supported for F16 and F32 models")
+					}
+
+					fn(api.ProgressResponse{Status: fmt.Sprintf("quantizing %s model to %s", filetype, quantization)})
+
+					blob, err := GetBlobsPath(baseLayer.Digest)
+					if err != nil {
+						return err
+					}
+
+					temp, err := os.CreateTemp(filepath.Dir(blob), quantization)
+					if err != nil {
+						return err
+					}
+					defer temp.Close()
+					defer os.Remove(temp.Name())
+
+					if err := llm.Quantize(blob, temp.Name(), ftype); err != nil {
+						return err
+					}
+
+					baseLayer.Layer, err = NewLayer(temp, baseLayer.Layer.MediaType)
+					if err != nil {
+						return err
+					}
 				}

-				deleteMap[manifest.Config.Digest] = struct{}{}
-				continue
+				if baseLayer.GGML != nil {
+					config.ModelFormat = cmp.Or(config.ModelFormat, baseLayer.GGML.Name())
+					config.ModelFamily = cmp.Or(config.ModelFamily, baseLayer.GGML.KV().Architecture())
+					config.ModelType = cmp.Or(config.ModelType, format.HumanNumber(baseLayer.GGML.KV().ParameterCount()))
+					config.FileType = cmp.Or(config.FileType, baseLayer.GGML.KV().FileType())
+					config.ModelFamilies = append(config.ModelFamilies, baseLayer.GGML.KV().Architecture())
+				}
+
+				layers = append(layers, baseLayer.Layer)
 			}
-			defer bin.Close()
-
-			var offset int64
-			for {
-				fn(api.ProgressResponse{Status: "creating model layer"})
-				if _, err := bin.Seek(offset, io.SeekStart); err != nil {
-					return err
-				}
-
-				ggml, size, err := llm.DecodeGGML(bin)
-				if errors.Is(err, io.EOF) {
-					break
-				} else if errors.Is(err, llm.ErrUnsupportedFormat) {
-					return fmt.Errorf("model binary specified in FROM field is not a valid gguf format model, %w", err)
-				} else if err != nil {
-					return err
-				}
-
-				config.SetModelFormat(ggml.Name())
-				config.SetModelFamily(ggml.KV().Architecture())
-				config.SetModelType(format.HumanNumber(ggml.KV().ParameterCount()))
-				config.SetFileType(ggml.KV().FileType())
-
-				mediatype := mediatype
-				if ggml.KV().Architecture() == "clip" {
-					mediatype = "application/vnd.ollama.image.projector"
-				}
-
-				sr := io.NewSectionReader(bin, offset, size)
-				layer, err := NewLayer(sr, mediatype)
-				if err != nil {
-					return err
-				}
-
-				layers.Add(layer)
-
-				offset += size
-			}
-		case "adapter":
-			if strings.HasPrefix(c.Args, "@") {
-				blobPath, err := GetBlobsPath(strings.TrimPrefix(c.Args, "@"))
-				if err != nil {
-					return err
-				}
-
-				c.Args = blobPath
-			}
-
-			fn(api.ProgressResponse{Status: "creating adapter layer"})
-			bin, err := os.Open(realpath(modelFileDir, c.Args))
-			if err != nil {
-				return err
-			}
-			defer bin.Close()
-
-			_, size, err := llm.DecodeGGML(bin)
+		case "license", "template", "system":
+			blob := strings.NewReader(c.Args)
+			layer, err := NewLayer(blob, mediatype)
 			if err != nil {
 				return err
 			}

-			sr := io.NewSectionReader(bin, 0, size)
-			layer, err := NewLayer(sr, mediatype)
-			if err != nil {
-				return err
+			if c.Name != "license" {
+				// replace
+				layers = slices.DeleteFunc(layers, func(layer *Layer) bool {
+					return layer.MediaType == mediatype
+				})
 			}

-			layers.Add(layer)
-		case "license":
-			fn(api.ProgressResponse{Status: "creating license layer"})
-
-			bin := strings.NewReader(c.Args)
-			layer, err := NewLayer(bin, mediatype)
-			if err != nil {
-				return err
-			}
-
-			layers.Add(layer)
-		case "template", "system":
-			fn(api.ProgressResponse{Status: fmt.Sprintf("creating %s layer", c.Name)})
-
-			bin := strings.NewReader(c.Args)
-			layer, err := NewLayer(bin, mediatype)
-			if err != nil {
-				return err
-			}
-
-			layers.Replace(layer)
+			layers = append(layers, layer)
 		case "message":
-			messages = append(messages, c.Args)
+			role, content, ok := strings.Cut(c.Args, ": ")
+			if !ok {
+				return fmt.Errorf("invalid message: %s", c.Args)
+			}
+
+			messages = append(messages, &api.Message{Role: role, Content: content})
 		default:
-			params[c.Name] = append(params[c.Name], c.Args)
+			ps, err := api.FormatParams(map[string][]string{c.Name: {c.Args}})
+			if err != nil {
+				return err
+			}
+
+			for k, v := range ps {
+				if ks, ok := parameters[k].([]string); ok {
+					parameters[k] = append(ks, v.([]string)...)
+				} else if vs, ok := v.([]string); ok {
+					parameters[k] = vs
+				} else {
+					parameters[k] = v
+				}
+			}
 		}
 	}

-	if len(messages) > 0 {
-		fn(api.ProgressResponse{Status: "creating parameters layer"})
+	var err2 error
+	layers = slices.DeleteFunc(layers, func(layer *Layer) bool {
+		switch layer.MediaType {
+		case "application/vnd.ollama.image.message":
+			// if there are new messages, remove the inherited ones
+			if len(messages) > 0 {
+				return true
+			}

-		msgs := make([]api.Message, 0)
+			return false
+		case "application/vnd.ollama.image.params":
+			// merge inherited parameters with new ones
+			r, err := layer.Open()
+			if err != nil {
+				err2 = err
+				return false
+			}
+			defer r.Close()

-		for _, m := range messages {
-			// todo: handle images
-			msg := strings.SplitN(m, ": ", 2)
-			msgs = append(msgs, api.Message{Role: msg[0], Content: msg[1]})
+			var ps map[string]any
+			if err := json.NewDecoder(r).Decode(&ps); err != nil {
+				err2 = err
+				return false
+			}
+
+			for k, v := range ps {
+				if _, ok := parameters[k]; !ok {
+					parameters[k] = v
+				}
+			}
+
+			return true
+		default:
+			return false
 		}
+	})

+	if err2 != nil {
+		return err2
+	}
+
+	if len(messages) > 0 {
 		var b bytes.Buffer
-		if err := json.NewEncoder(&b).Encode(msgs); err != nil {
+		if err := json.NewEncoder(&b).Encode(messages); err != nil {
 			return err
 		}

@@ -556,39 +516,25 @@ func CreateModel(ctx context.Context, name, modelFileDir, quantization string, c
 			return err
 		}

-		layers.Replace(layer)
+		layers = append(layers, layer)
 	}

-	if len(params) > 0 {
-		fn(api.ProgressResponse{Status: "creating parameters layer"})
-
-		formattedParams, err := api.FormatParams(params)
-		if err != nil {
-			return err
-		}
-
-		for k, v := range fromParams {
-			if _, ok := formattedParams[k]; !ok {
-				formattedParams[k] = v
-			}
-		}
-
+	if len(parameters) > 0 {
 		var b bytes.Buffer
-		if err := json.NewEncoder(&b).Encode(formattedParams); err != nil {
+		if err := json.NewEncoder(&b).Encode(parameters); err != nil {
 			return err
 		}

-		fn(api.ProgressResponse{Status: "creating config layer"})
 		layer, err := NewLayer(&b, "application/vnd.ollama.image.params")
 		if err != nil {
 			return err
 		}

-		layers.Replace(layer)
+		layers = append(layers, layer)
 	}

-	digests := make([]string, len(layers.items))
-	for i, layer := range layers.items {
+	digests := make([]string, len(layers))
+	for i, layer := range layers {
 		digests[i] = layer.Digest
 	}

@@ -599,36 +545,38 @@ func CreateModel(ctx context.Context, name, modelFileDir, quantization string, c
 		return err
 	}

-	configLayer, err := NewLayer(&b, "application/vnd.docker.container.image.v1+json")
+	layer, err := NewLayer(&b, "application/vnd.docker.container.image.v1+json")
 	if err != nil {
 		return err
 	}

-	delete(deleteMap, configLayer.Digest)
+	for _, layer := range append(layers, layer) {
+		if layer.status != "" {
+			fn(api.ProgressResponse{Status: layer.status})
+		}
+	}

-	for _, layer := range append(layers.items, configLayer) {
-		committed, err := layer.Commit()
-		if err != nil {
-			return err
+	unref := make(map[string]struct{})
+	if manifest, _, err := GetManifest(ParseModelPath(name)); err == nil {
+		for _, layer := range manifest.Layers {
+			if !slices.Contains(digests, layer.Digest) {
+				unref[layer.Digest] = struct{}{}
+			}
 		}

-		status := "writing layer"
-		if !committed {
-			status = "using already created layer"
+		if manifest.Config.Digest != layer.Digest {
+			unref[manifest.Config.Digest] = struct{}{}
 		}
-
-		fn(api.ProgressResponse{Status: fmt.Sprintf("%s %s", status, layer.Digest)})
-
-		delete(deleteMap, layer.Digest)
 	}

 	fn(api.ProgressResponse{Status: "writing manifest"})
-	if err := WriteManifest(name, configLayer, layers.items); err != nil {
+	if err := WriteManifest(name, layer, layers); err != nil {
 		return err
 	}

-	if noprune := os.Getenv("OLLAMA_NOPRUNE"); noprune == "" {
-		if err := deleteUnusedLayers(nil, deleteMap, false); err != nil {
+	if os.Getenv("OLLAMA_NOPRUNE") == "" && len(unref) > 0 {
+		fn(api.ProgressResponse{Status: "removing unused layers"})
+		if err := deleteUnusedLayers(nil, unref, false); err != nil {
 			return err
 		}
 	}
@@ -637,74 +585,6 @@ func CreateModel(ctx context.Context, name, modelFileDir, quantization string, c
 	return nil
 }

-func convertModel(name, path string, fn func(resp api.ProgressResponse)) (string, error) {
-	r, err := zip.OpenReader(path)
-	if err != nil {
-		return "", err
-	}
-	defer r.Close()
-
-	tempDir, err := os.MkdirTemp("", "ollama-convert")
-	if err != nil {
-		return "", err
-	}
-	defer os.RemoveAll(tempDir)
-
-	fn(api.ProgressResponse{Status: "unpacking model metadata"})
-	for _, f := range r.File {
-		fpath := filepath.Join(tempDir, f.Name)
-		outFile, err := os.OpenFile(fpath, os.O_WRONLY|os.O_CREATE|os.O_TRUNC, f.Mode())
-		if err != nil {
-			return "", err
-		}
-
-		rc, err := f.Open()
-		if err != nil {
-			return "", err
-		}
-
-		_, err = io.Copy(outFile, rc)
-		if err != nil {
-			return "", err
-		}
-
-		outFile.Close()
-		rc.Close()
-	}
-
-	mf, err := convert.GetModelFormat(tempDir)
-	if err != nil {
-		return "", err
-	}
-
-	params, err := mf.GetParams(tempDir)
-	if err != nil {
-		return "", err
-	}
-
-	mArch, err := mf.GetModelArch(name, tempDir, params)
-	if err != nil {
-		return "", err
-	}
-
-	fn(api.ProgressResponse{Status: "processing tensors"})
-	if err := mArch.GetTensors(); err != nil {
-		return "", err
-	}
-
-	if err := mArch.LoadVocab(); err != nil {
-		return "", err
-	}
-
-	fn(api.ProgressResponse{Status: "converting model"})
-	path, err = mArch.WriteGGUF()
-	if err != nil {
-		return "", err
-	}
-
-	return path, nil
-}
-
 func CopyModel(src, dst model.Name) error {
 	if !dst.IsFullyQualified() {
 		return model.Unqualified(dst)
@@ -901,67 +781,6 @@ func DeleteModel(name string) error {
 	return nil
 }

-func ShowModelfile(model *Model) (string, error) {
-	var mt struct {
-		*Model
-		From       string
-		Parameters map[string][]any
-	}
-
-	mt.Parameters = make(map[string][]any)
-	for k, v := range model.Options {
-		if s, ok := v.([]any); ok {
-			mt.Parameters[k] = s
-			continue
-		}
-
-		mt.Parameters[k] = []any{v}
-	}
-
-	mt.Model = model
-	mt.From = model.ModelPath
-
-	if model.ParentModel != "" {
-		mt.From = model.ParentModel
-	}
-
-	modelFile := `# Modelfile generated by "ollama show"
-# To build a new Modelfile based on this one, replace the FROM line with:
-# FROM {{ .ShortName }}
-
-FROM {{ .From }}
-TEMPLATE """{{ .Template }}"""
-
-{{- if .System }}
-SYSTEM """{{ .System }}"""
-{{- end }}
-
-{{- range $adapter := .AdapterPaths }}
-ADAPTER {{ $adapter }}
-{{- end }}
-
-{{- range $k, $v := .Parameters }}
-{{- range $parameter := $v }}
-PARAMETER {{ $k }} {{ printf "%#v" $parameter }}
-{{- end }}
-{{- end }}`
-
-	tmpl, err := template.New("").Parse(modelFile)
-	if err != nil {
-		slog.Info(fmt.Sprintf("error parsing template: %q", err))
-		return "", err
-	}
-
-	var buf bytes.Buffer
-
-	if err = tmpl.Execute(&buf, mt); err != nil {
-		slog.Info(fmt.Sprintf("error executing template: %q", err))
-		return "", err
-	}
-
-	return buf.String(), nil
-}
-
 func PushModel(ctx context.Context, name string, regOpts *registryOptions, fn func(api.ProgressResponse)) error {
 	mp := ParseModelPath(name)
 	fn(api.ProgressResponse{Status: "retrieving manifest"})
--- a/server/layers.go
+++ b/server/layers.go
@@ -5,39 +5,14 @@ import (
 	"fmt"
 	"io"
 	"os"
-	"strings"
-
-	"golang.org/x/exp/slices"
 )

-type Layers struct {
-	items []*Layer
-}
-
-func (ls *Layers) Add(layer *Layer) {
-	if layer.Size > 0 {
-		ls.items = append(ls.items, layer)
-	}
-}
-
-func (ls *Layers) Replace(layer *Layer) {
-	if layer.Size > 0 {
-		mediatype := layer.MediaType
-		layers := slices.DeleteFunc(ls.items, func(l *Layer) bool {
-			return l.MediaType == mediatype
-		})
-
-		ls.items = append(layers, layer)
-	}
-}
-
 type Layer struct {
 	MediaType string `json:"mediaType"`
 	Digest    string `json:"digest"`
 	Size      int64  `json:"size"`
 	From      string `json:"from,omitempty"`
-
-	tempFileName string
+	status    string
 }

 func NewLayer(r io.Reader, mediatype string) (*Layer, error) {
@@ -46,14 +21,12 @@ func NewLayer(r io.Reader, mediatype string) (*Layer, error) {
 		return nil, err
 	}

-	const delimiter = "-"
-
-	pattern := strings.Join([]string{"sha256", "*-partial"}, delimiter)
-	temp, err := os.CreateTemp(blobs, pattern)
+	temp, err := os.CreateTemp(blobs, "sha256-")
 	if err != nil {
 		return nil, err
 	}
 	defer temp.Close()
+	defer os.Remove(temp.Name())

 	sha256sum := sha256.New()
 	n, err := io.Copy(io.MultiWriter(temp, sha256sum), r)
@@ -61,11 +34,29 @@ func NewLayer(r io.Reader, mediatype string) (*Layer, error) {
 		return nil, err
 	}

+	if err := temp.Close(); err != nil {
+		return nil, err
+	}
+
+	digest := fmt.Sprintf("sha256:%x", sha256sum.Sum(nil))
+	blob, err := GetBlobsPath(digest)
+	if err != nil {
+		return nil, err
+	}
+
+	status := "using existing layer"
+	if _, err := os.Stat(blob); err != nil {
+		status = "creating new layer"
+		if err := os.Rename(temp.Name(), blob); err != nil {
+			return nil, err
+		}
+	}
+
 	return &Layer{
-		MediaType:    mediatype,
-		Digest:       fmt.Sprintf("sha256:%x", sha256sum.Sum(nil)),
-		Size:         n,
-		tempFileName: temp.Name(),
+		MediaType: mediatype,
+		Digest:    digest,
+		Size:      n,
+		status:    fmt.Sprintf("%s %s", status, digest),
 	}, nil
 }

@@ -85,21 +76,15 @@ func NewLayerFromLayer(digest, mediatype, from string) (*Layer, error) {
 		Digest:    digest,
 		Size:      fi.Size(),
 		From:      from,
+		status:    fmt.Sprintf("using existing layer %s", digest),
 	}, nil
 }

-func (l *Layer) Commit() (bool, error) {
-	// always remove temp
-	defer os.Remove(l.tempFileName)
-
+func (l *Layer) Open() (io.ReadCloser, error) {
 	blob, err := GetBlobsPath(l.Digest)
 	if err != nil {
-		return false, err
+		return nil, err
 	}

-	if _, err := os.Stat(blob); err != nil {
-		return true, os.Rename(l.tempFileName, blob)
-	}
-
-	return false, nil
+	return os.Open(blob)
 }
--- a/server/model.go
+++ b/server/model.go
@@ -0,0 +1,259 @@
+package server
+
+import (
+	"archive/zip"
+	"bytes"
+	"context"
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+	"os"
+	"path/filepath"
+
+	"github.com/ollama/ollama/api"
+	"github.com/ollama/ollama/convert"
+	"github.com/ollama/ollama/llm"
+	"github.com/ollama/ollama/types/model"
+)
+
+type layerWithGGML struct {
+	*Layer
+	*llm.GGML
+}
+
+func parseFromModel(ctx context.Context, name model.Name, fn func(api.ProgressResponse)) (layers []*layerWithGGML, err error) {
+	modelpath := ParseModelPath(name.String())
+	manifest, _, err := GetManifest(modelpath)
+	switch {
+	case errors.Is(err, os.ErrNotExist):
+		if err := PullModel(ctx, name.String(), &registryOptions{}, fn); err != nil {
+			return nil, err
+		}
+
+		modelpath = ParseModelPath(name.String())
+		manifest, _, err = GetManifest(modelpath)
+		if err != nil {
+			return nil, err
+		}
+	case err != nil:
+		return nil, err
+	}
+
+	for _, layer := range manifest.Layers {
+		layer, err := NewLayerFromLayer(layer.Digest, layer.MediaType, modelpath.GetShortTagname())
+		if err != nil {
+			return nil, err
+		}
+
+		switch layer.MediaType {
+		case "application/vnd.ollama.image.model",
+			"application/vnd.ollama.image.projector",
+			"application/vnd.ollama.image.adapter":
+			blobpath, err := GetBlobsPath(layer.Digest)
+			if err != nil {
+				return nil, err
+			}
+
+			blob, err := os.Open(blobpath)
+			if err != nil {
+				return nil, err
+			}
+			defer blob.Close()
+
+			ggml, _, err := llm.DecodeGGML(blob)
+			if err != nil {
+				return nil, err
+			}
+
+			layers = append(layers, &layerWithGGML{layer, ggml})
+		default:
+			layers = append(layers, &layerWithGGML{layer, nil})
+		}
+
+	}
+
+	return layers, nil
+}
+
+func parseFromZipFile(_ context.Context, file *os.File, fn func(api.ProgressResponse)) (layers []*layerWithGGML, err error) {
+	stat, err := file.Stat()
+	if err != nil {
+		return nil, err
+	}
+
+	r, err := zip.NewReader(file, stat.Size())
+	if err != nil {
+		return nil, err
+	}
+
+	tempdir, err := os.MkdirTemp(filepath.Dir(file.Name()), "")
+	if err != nil {
+		return nil, err
+	}
+	defer os.RemoveAll(tempdir)
+
+	fn(api.ProgressResponse{Status: "unpacking model metadata"})
+	for _, f := range r.File {
+		// TODO(mxyng): this should not write out all files to disk
+		outfile, err := os.Create(filepath.Join(tempdir, f.Name))
+		if err != nil {
+			return nil, err
+		}
+
+		infile, err := f.Open()
+		if err != nil {
+			return nil, err
+		}
+
+		if _, err = io.Copy(outfile, infile); err != nil {
+			return nil, err
+		}
+
+		if err := outfile.Close(); err != nil {
+			return nil, err
+		}
+
+		if err := infile.Close(); err != nil {
+			return nil, err
+		}
+	}
+
+	mf, err := convert.GetModelFormat(tempdir)
+	if err != nil {
+		return nil, err
+	}
+
+	params, err := mf.GetParams(tempdir)
+	if err != nil {
+		return nil, err
+	}
+
+	mArch, err := mf.GetModelArch("", tempdir, params)
+	if err != nil {
+		return nil, err
+	}
+
+	fn(api.ProgressResponse{Status: "processing tensors"})
+	if err := mArch.GetTensors(); err != nil {
+		return nil, err
+	}
+
+	if err := mArch.LoadVocab(); err != nil {
+		return nil, err
+	}
+
+	fn(api.ProgressResponse{Status: "converting model"})
+
+	// TODO(mxyng): this should write directly into a layer
+	// e.g. NewLayer(arch.Reader(), "application/vnd.ollama.image.model")
+	temp, err := os.CreateTemp(tempdir, "fp16")
+	if err != nil {
+		return nil, err
+	}
+	defer temp.Close()
+	defer os.Remove(temp.Name())
+
+	if err = mArch.WriteGGUF(temp); err != nil {
+		return nil, err
+	}
+
+	if _, err := temp.Seek(0, io.SeekStart); err != nil {
+		return nil, err
+	}
+
+	layer, err := NewLayer(temp, "application/vnd.ollama.image.model")
+	if err != nil {
+		return nil, fmt.Errorf("aaa: %w", err)
+	}
+
+	blobpath, err := GetBlobsPath(layer.Digest)
+	if err != nil {
+		return nil, err
+	}
+
+	bin, err := os.Open(blobpath)
+	if err != nil {
+		return nil, err
+	}
+	defer bin.Close()
+
+	ggml, _, err := llm.DecodeGGML(bin)
+	if err != nil {
+		return nil, err
+	}
+
+	layer, err = NewLayerFromLayer(layer.Digest, layer.MediaType, "")
+	if err != nil {
+		return nil, err
+	}
+
+	layers = append(layers, &layerWithGGML{layer, ggml})
+	return layers, nil
+}
+
+func parseFromFile(ctx context.Context, file *os.File, fn func(api.ProgressResponse)) (layers []*layerWithGGML, err error) {
+	sr := io.NewSectionReader(file, 0, 512)
+	contentType, err := detectContentType(sr)
+	if err != nil {
+		return nil, err
+	}
+
+	switch contentType {
+	case "gguf", "ggla":
+		// noop
+	case "application/zip":
+		return parseFromZipFile(ctx, file, fn)
+	default:
+		return nil, fmt.Errorf("unsupported content type: %s", contentType)
+	}
+
+	stat, err := file.Stat()
+	if err != nil {
+		return nil, err
+	}
+
+	var offset int64
+	for offset < stat.Size() {
+		ggml, n, err := llm.DecodeGGML(file)
+		if errors.Is(err, io.EOF) {
+			break
+		} else if err != nil {
+			return nil, err
+		}
+
+		mediatype := "application/vnd.ollama.image.model"
+		if ggml.Name() == "ggla" {
+			mediatype = "application/vnd.ollama.image.adapter"
+		} else if ggml.KV().Architecture() == "clip" {
+			mediatype = "application/vnd.ollama.image.projector"
+		}
+
+		layer, err := NewLayer(io.NewSectionReader(file, offset, n), mediatype)
+		if err != nil {
+			return nil, err
+		}
+
+		layers = append(layers, &layerWithGGML{layer, ggml})
+		offset = n
+	}
+
+	return layers, nil
+}
+
+func detectContentType(r io.Reader) (string, error) {
+	var b bytes.Buffer
+	if _, err := io.Copy(&b, r); err != nil {
+		return "", err
+	}
+
+	if contentType := llm.DetectGGMLType(b.Bytes()); contentType != "" {
+		return contentType, nil
+	}
+
+	if contentType := http.DetectContentType(b.Bytes()); contentType != "application/octet-stream" {
+		return contentType, nil
+	}
+
+	return "unknown", nil
+}
--- a/server/routes.go
+++ b/server/routes.go
@@ -1,6 +1,7 @@
 package server

 import (
+	"cmp"
 	"context"
 	"encoding/json"
 	"errors"
@@ -28,7 +29,6 @@ import (
 	"github.com/ollama/ollama/gpu"
 	"github.com/ollama/ollama/llm"
 	"github.com/ollama/ollama/openai"
-	"github.com/ollama/ollama/parser"
 	"github.com/ollama/ollama/types/model"
 	"github.com/ollama/ollama/version"
 )
@@ -522,28 +522,17 @@ func (s *Server) PushModelHandler(c *gin.Context) {

 func (s *Server) CreateModelHandler(c *gin.Context) {
 	var req api.CreateRequest
-	err := c.ShouldBindJSON(&req)
-	switch {
-	case errors.Is(err, io.EOF):
+	if err := c.ShouldBindJSON(&req); errors.Is(err, io.EOF) {
 		c.AbortWithStatusJSON(http.StatusBadRequest, gin.H{"error": "missing request body"})
 		return
-	case err != nil:
+	} else if err != nil {
 		c.AbortWithStatusJSON(http.StatusBadRequest, gin.H{"error": err.Error()})
 		return
 	}

-	var model string
-	if req.Model != "" {
-		model = req.Model
-	} else if req.Name != "" {
-		model = req.Name
-	} else {
-		c.AbortWithStatusJSON(http.StatusBadRequest, gin.H{"error": "model is required"})
-		return
-	}
-
-	if err := ParseModelPath(model).Validate(); err != nil {
-		c.AbortWithStatusJSON(http.StatusBadRequest, gin.H{"error": err.Error()})
+	name := model.ParseName(cmp.Or(req.Model, req.Name))
+	if !name.IsValid() {
+		c.AbortWithStatusJSON(http.StatusBadRequest, gin.H{"error": "invalid model name"})
 		return
 	}

@@ -552,19 +541,19 @@ func (s *Server) CreateModelHandler(c *gin.Context) {
 		return
 	}

-	var modelfile io.Reader = strings.NewReader(req.Modelfile)
+	var r io.Reader = strings.NewReader(req.Modelfile)
 	if req.Path != "" && req.Modelfile == "" {
-		mf, err := os.Open(req.Path)
+		f, err := os.Open(req.Path)
 		if err != nil {
 			c.AbortWithStatusJSON(http.StatusBadRequest, gin.H{"error": fmt.Sprintf("error reading modelfile: %s", err)})
 			return
 		}
-		defer mf.Close()
+		defer f.Close()

-		modelfile = mf
+		r = f
 	}

-	commands, err := parser.Parse(modelfile)
+	modelfile, err := model.ParseFile(r)
 	if err != nil {
 		c.AbortWithStatusJSON(http.StatusBadRequest, gin.H{"error": err.Error()})
 		return
@@ -580,7 +569,7 @@ func (s *Server) CreateModelHandler(c *gin.Context) {
 		ctx, cancel := context.WithCancel(c.Request.Context())
 		defer cancel()

-		if err := CreateModel(ctx, model, filepath.Dir(req.Path), req.Quantization, commands, fn); err != nil {
+		if err := CreateModel(ctx, name.String(), filepath.Dir(req.Path), strings.ToUpper(req.Quantization), modelfile, fn); err != nil {
 			ch <- gin.H{"error": err.Error()}
 		}
 	}()
@@ -728,12 +717,12 @@ func GetModelInfo(req api.ShowRequest) (*api.ShowResponse, error) {
 		}
 	}

-	mf, err := ShowModelfile(model)
-	if err != nil {
-		return nil, err
-	}
-
-	resp.Modelfile = mf
+	var sb strings.Builder
+	fmt.Fprintln(&sb, "# Modelfile generate by \"ollama show\"")
+	fmt.Fprintln(&sb, "# To build a new Modelfile based on this, replace FROM with:")
+	fmt.Fprintf(&sb, "# FROM %s\n\n", model.ShortName)
+	fmt.Fprint(&sb, model.String())
+	resp.Modelfile = sb.String()

 	return resp, nil
 }
@@ -872,11 +861,6 @@ func (s *Server) CreateBlobHandler(c *gin.Context) {
 		return
 	}

-	if _, err := layer.Commit(); err != nil {
-		c.AbortWithStatusJSON(http.StatusInternalServerError, gin.H{"error": err.Error()})
-		return
-	}
-
 	c.Status(http.StatusCreated)
 }

--- a/server/routes_test.go
+++ b/server/routes_test.go
@@ -17,7 +17,7 @@ import (
 	"github.com/stretchr/testify/assert"

 	"github.com/ollama/ollama/api"
-	"github.com/ollama/ollama/parser"
+	"github.com/ollama/ollama/types/model"
 	"github.com/ollama/ollama/version"
 )

@@ -55,13 +55,13 @@ func Test_Routes(t *testing.T) {
 	createTestModel := func(t *testing.T, name string) {
 		fname := createTestFile(t, "ollama-model")

-		modelfile := strings.NewReader(fmt.Sprintf("FROM %s\nPARAMETER seed 42\nPARAMETER top_p 0.9\nPARAMETER stop foo\nPARAMETER stop bar", fname))
-		commands, err := parser.Parse(modelfile)
+		r := strings.NewReader(fmt.Sprintf("FROM %s\nPARAMETER seed 42\nPARAMETER top_p 0.9\nPARAMETER stop foo\nPARAMETER stop bar", fname))
+		modelfile, err := model.ParseFile(r)
 		assert.Nil(t, err)
 		fn := func(resp api.ProgressResponse) {
 			t.Logf("Status: %s", resp.Status)
 		}
-		err = CreateModel(context.TODO(), name, "", "", commands, fn)
+		err = CreateModel(context.TODO(), name, "", "", modelfile, fn)
 		assert.Nil(t, err)
 	}

@@ -124,14 +124,12 @@ func Test_Routes(t *testing.T) {
 			Method: http.MethodPost,
 			Path:   "/api/create",
 			Setup: func(t *testing.T, req *http.Request) {
-				f, err := os.CreateTemp(t.TempDir(), "ollama-model")
-				assert.Nil(t, err)
-				defer f.Close()
+				fname := createTestFile(t, "ollama-model")

 				stream := false
 				createReq := api.CreateRequest{
 					Name:      "t-bone",
-					Modelfile: fmt.Sprintf("FROM %s", f.Name()),
+					Modelfile: fmt.Sprintf("FROM %s", fname),
 					Stream:    &stream,
 				}
 				jsonData, err := json.Marshal(createReq)
@@ -216,28 +214,25 @@ func Test_Routes(t *testing.T) {
 	httpSrv := httptest.NewServer(router)
 	t.Cleanup(httpSrv.Close)

-	workDir, err := os.MkdirTemp("", "ollama-test")
-	assert.Nil(t, err)
-	defer os.RemoveAll(workDir)
-	os.Setenv("OLLAMA_MODELS", workDir)
+	t.Setenv("OLLAMA_MODELS", t.TempDir())

 	for _, tc := range testCases {
-		t.Logf("Running Test: [%s]", tc.Name)
-		u := httpSrv.URL + tc.Path
-		req, err := http.NewRequestWithContext(context.TODO(), tc.Method, u, nil)
-		assert.Nil(t, err)
+		t.Run(tc.Name, func(t *testing.T) {
+			u := httpSrv.URL + tc.Path
+			req, err := http.NewRequestWithContext(context.TODO(), tc.Method, u, nil)
+			assert.Nil(t, err)

-		if tc.Setup != nil {
-			tc.Setup(t, req)
-		}
+			if tc.Setup != nil {
+				tc.Setup(t, req)
+			}

-		resp, err := httpSrv.Client().Do(req)
-		assert.Nil(t, err)
-		defer resp.Body.Close()
-
-		if tc.Expected != nil {
-			tc.Expected(t, resp)
-		}
+			resp, err := httpSrv.Client().Do(req)
+			assert.Nil(t, err)
+			defer resp.Body.Close()

+			if tc.Expected != nil {
+				tc.Expected(t, resp)
+			}
+		})
 	}
 }
--- a/server/sched_test.go
+++ b/server/sched_test.go
@@ -58,7 +58,7 @@ func TestInitScheduler(t *testing.T) {
 }

 func TestLoad(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 5*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 20*time.Millisecond)
 	defer done()
 	s := InitScheduler(ctx)
 	var ggml *llm.GGML // value not used in tests
@@ -174,7 +174,7 @@ func newScenario(t *testing.T, ctx context.Context, modelName string, estimatedV
 }

 func TestRequests(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 100*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 500*time.Millisecond)
 	defer done()

 	// Same model, same request
@@ -329,7 +329,7 @@ func TestRequests(t *testing.T) {
 }

 func TestGetRunner(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 20*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 100*time.Millisecond)
 	defer done()

 	// Same model, same request
@@ -391,7 +391,7 @@ func TestGetRunner(t *testing.T) {

 // TODO - add one scenario that triggers the bogus finished event with positive ref count
 func TestPrematureExpired(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 100*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 500*time.Millisecond)
 	defer done()

 	// Same model, same request
@@ -436,7 +436,7 @@ func TestPrematureExpired(t *testing.T) {
 }

 func TestUseLoadedRunner(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 5*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 100*time.Millisecond)
 	req := &LlmRequest{
 		ctx:             ctx,
 		opts:            api.DefaultOptions(),
@@ -461,7 +461,7 @@ func TestUseLoadedRunner(t *testing.T) {
 }

 func TestUpdateFreeSpace(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 5*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 100*time.Millisecond)
 	defer done()
 	gpus := gpu.GpuInfoList{
 		{
@@ -494,7 +494,7 @@ func TestUpdateFreeSpace(t *testing.T) {
 }

 func TestFindRunnerToUnload(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 5*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 100*time.Millisecond)
 	defer done()
 	req := &LlmRequest{
 		ctx:  ctx,
@@ -518,7 +518,7 @@ func TestFindRunnerToUnload(t *testing.T) {
 }

 func TestNeedsReload(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 5*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 100*time.Millisecond)
 	defer done()

 	llm := &mockLlm{}
@@ -562,7 +562,7 @@ func TestNeedsReload(t *testing.T) {
 }

 func TestUnloadAllRunners(t *testing.T) {
-	ctx, done := context.WithTimeout(context.Background(), 5*time.Millisecond)
+	ctx, done := context.WithTimeout(context.Background(), 100*time.Millisecond)
 	defer done()

 	llm1 := &mockLlm{}
--- a/types/model/file.go
+++ b/types/model/file.go
@@ -0,0 +1,303 @@
+package model
+
+import (
+	"bufio"
+	"bytes"
+	"errors"
+	"fmt"
+	"io"
+	"strconv"
+	"strings"
+)
+
+type File struct {
+	Commands []Command
+}
+
+func (f File) String() string {
+	var sb strings.Builder
+	for _, cmd := range f.Commands {
+		fmt.Fprintln(&sb, cmd.String())
+	}
+
+	return sb.String()
+}
+
+type Command struct {
+	Name string
+	Args string
+}
+
+func (c Command) String() string {
+	var sb strings.Builder
+	switch c.Name {
+	case "model":
+		fmt.Fprintf(&sb, "FROM %s", c.Args)
+	case "license", "template", "system", "adapter":
+		fmt.Fprintf(&sb, "%s %s", strings.ToUpper(c.Name), quote(c.Args))
+	case "message":
+		role, message, _ := strings.Cut(c.Args, ": ")
+		fmt.Fprintf(&sb, "MESSAGE %s %s", role, quote(message))
+	default:
+		fmt.Fprintf(&sb, "PARAMETER %s %s", c.Name, quote(c.Args))
+	}
+
+	return sb.String()
+}
+
+type state int
+
+const (
+	stateNil state = iota
+	stateName
+	stateValue
+	stateParameter
+	stateMessage
+	stateComment
+)
+
+var (
+	errMissingFrom        = errors.New("no FROM line")
+	errInvalidMessageRole = errors.New("message role must be one of \"system\", \"user\", or \"assistant\"")
+	errInvalidCommand     = errors.New("command must be one of \"from\", \"license\", \"template\", \"system\", \"adapter\", \"parameter\", or \"message\"")
+)
+
+func ParseFile(r io.Reader) (*File, error) {
+	var cmd Command
+	var curr state
+	var b bytes.Buffer
+	var role string
+
+	var f File
+
+	br := bufio.NewReader(r)
+	for {
+		r, _, err := br.ReadRune()
+		if errors.Is(err, io.EOF) {
+			break
+		} else if err != nil {
+			return nil, err
+		}
+
+		next, r, err := parseRuneForState(r, curr)
+		if errors.Is(err, io.ErrUnexpectedEOF) {
+			return nil, fmt.Errorf("%w: %s", err, b.String())
+		} else if err != nil {
+			return nil, err
+		}
+
+		// process the state transition, some transitions need to be intercepted and redirected
+		if next != curr {
+			switch curr {
+			case stateName:
+				if !isValidCommand(b.String()) {
+					return nil, errInvalidCommand
+				}
+
+				// next state sometimes depends on the current buffer value
+				switch s := strings.ToLower(b.String()); s {
+				case "from":
+					cmd.Name = "model"
+				case "parameter":
+					// transition to stateParameter which sets command name
+					next = stateParameter
+				case "message":
+					// transition to stateMessage which validates the message role
+					next = stateMessage
+					fallthrough
+				default:
+					cmd.Name = s
+				}
+			case stateParameter:
+				cmd.Name = b.String()
+			case stateMessage:
+				if !isValidMessageRole(b.String()) {
+					return nil, errInvalidMessageRole
+				}
+
+				role = b.String()
+			case stateComment, stateNil:
+				// pass
+			case stateValue:
+				s, ok := unquote(b.String())
+				if !ok || isSpace(r) {
+					if _, err := b.WriteRune(r); err != nil {
+						return nil, err
+					}
+
+					continue
+				}
+
+				if role != "" {
+					s = role + ": " + s
+					role = ""
+				}
+
+				cmd.Args = s
+				f.Commands = append(f.Commands, cmd)
+			}
+
+			b.Reset()
+			curr = next
+		}
+
+		if strconv.IsPrint(r) {
+			if _, err := b.WriteRune(r); err != nil {
+				return nil, err
+			}
+		}
+	}
+
+	// flush the buffer
+	switch curr {
+	case stateComment, stateNil:
+		// pass; nothing to flush
+	case stateValue:
+		s, ok := unquote(b.String())
+		if !ok {
+			return nil, io.ErrUnexpectedEOF
+		}
+
+		if role != "" {
+			s = role + ": " + s
+		}
+
+		cmd.Args = s
+		f.Commands = append(f.Commands, cmd)
+	default:
+		return nil, io.ErrUnexpectedEOF
+	}
+
+	for _, cmd := range f.Commands {
+		if cmd.Name == "model" {
+			return &f, nil
+		}
+	}
+
+	return nil, errMissingFrom
+}
+
+func parseRuneForState(r rune, cs state) (state, rune, error) {
+	switch cs {
+	case stateNil:
+		switch {
+		case r == '#':
+			return stateComment, 0, nil
+		case isSpace(r), isNewline(r):
+			return stateNil, 0, nil
+		default:
+			return stateName, r, nil
+		}
+	case stateName:
+		switch {
+		case isAlpha(r):
+			return stateName, r, nil
+		case isSpace(r):
+			return stateValue, 0, nil
+		default:
+			return stateNil, 0, errInvalidCommand
+		}
+	case stateValue:
+		switch {
+		case isNewline(r):
+			return stateNil, r, nil
+		case isSpace(r):
+			return stateNil, r, nil
+		default:
+			return stateValue, r, nil
+		}
+	case stateParameter:
+		switch {
+		case isAlpha(r), isNumber(r), r == '_':
+			return stateParameter, r, nil
+		case isSpace(r):
+			return stateValue, 0, nil
+		default:
+			return stateNil, 0, io.ErrUnexpectedEOF
+		}
+	case stateMessage:
+		switch {
+		case isAlpha(r):
+			return stateMessage, r, nil
+		case isSpace(r):
+			return stateValue, 0, nil
+		default:
+			return stateNil, 0, io.ErrUnexpectedEOF
+		}
+	case stateComment:
+		switch {
+		case isNewline(r):
+			return stateNil, 0, nil
+		default:
+			return stateComment, 0, nil
+		}
+	default:
+		return stateNil, 0, errors.New("")
+	}
+}
+
+func quote(s string) string {
+	if strings.Contains(s, "\n") || strings.HasPrefix(s, " ") || strings.HasSuffix(s, " ") {
+		if strings.Contains(s, "\"") {
+			return `"""` + s + `"""`
+		}
+
+		return `"` + s + `"`
+	}
+
+	return s
+}
+
+func unquote(s string) (string, bool) {
+	if len(s) == 0 {
+		return "", false
+	}
+
+	// TODO: single quotes
+	if len(s) >= 3 && s[:3] == `"""` {
+		if len(s) >= 6 && s[len(s)-3:] == `"""` {
+			return s[3 : len(s)-3], true
+		}
+
+		return "", false
+	}
+
+	if len(s) >= 1 && s[0] == '"' {
+		if len(s) >= 2 && s[len(s)-1] == '"' {
+			return s[1 : len(s)-1], true
+		}
+
+		return "", false
+	}
+
+	return s, true
+}
+
+func isAlpha(r rune) bool {
+	return r >= 'a' && r <= 'z' || r >= 'A' && r <= 'Z'
+}
+
+func isNumber(r rune) bool {
+	return r >= '0' && r <= '9'
+}
+
+func isSpace(r rune) bool {
+	return r == ' ' || r == '\t'
+}
+
+func isNewline(r rune) bool {
+	return r == '\r' || r == '\n'
+}
+
+func isValidMessageRole(role string) bool {
+	return role == "system" || role == "user" || role == "assistant"
+}
+
+func isValidCommand(cmd string) bool {
+	switch strings.ToLower(cmd) {
+	case "from", "license", "template", "system", "adapter", "parameter", "message":
+		return true
+	default:
+		return false
+	}
+}
--- a/types/model/file_test.go
+++ b/types/model/file_test.go
@@ -0,0 +1,507 @@
+package model
+
+import (
+	"bytes"
+	"fmt"
+	"io"
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestParseFileFile(t *testing.T) {
+	input := `
+FROM model1
+ADAPTER adapter1
+LICENSE MIT
+PARAMETER param1 value1
+PARAMETER param2 value2
+TEMPLATE template1
+`
+
+	reader := strings.NewReader(input)
+
+	modelfile, err := ParseFile(reader)
+	assert.NoError(t, err)
+
+	expectedCommands := []Command{
+		{Name: "model", Args: "model1"},
+		{Name: "adapter", Args: "adapter1"},
+		{Name: "license", Args: "MIT"},
+		{Name: "param1", Args: "value1"},
+		{Name: "param2", Args: "value2"},
+		{Name: "template", Args: "template1"},
+	}
+
+	assert.Equal(t, expectedCommands, modelfile.Commands)
+}
+
+func TestParseFileFrom(t *testing.T) {
+	var cases = []struct {
+		input    string
+		expected []Command
+		err      error
+	}{
+		{
+			"FROM foo",
+			[]Command{{Name: "model", Args: "foo"}},
+			nil,
+		},
+		{
+			"FROM /path/to/model",
+			[]Command{{Name: "model", Args: "/path/to/model"}},
+			nil,
+		},
+		{
+			"FROM /path/to/model/fp16.bin",
+			[]Command{{Name: "model", Args: "/path/to/model/fp16.bin"}},
+			nil,
+		},
+		{
+			"FROM llama3:latest",
+			[]Command{{Name: "model", Args: "llama3:latest"}},
+			nil,
+		},
+		{
+			"FROM llama3:7b-instruct-q4_K_M",
+			[]Command{{Name: "model", Args: "llama3:7b-instruct-q4_K_M"}},
+			nil,
+		},
+		{
+			"", nil, errMissingFrom,
+		},
+		{
+			"PARAMETER param1 value1",
+			nil,
+			errMissingFrom,
+		},
+		{
+			"PARAMETER param1 value1\nFROM foo",
+			[]Command{{Name: "param1", Args: "value1"}, {Name: "model", Args: "foo"}},
+			nil,
+		},
+	}
+
+	for _, c := range cases {
+		t.Run("", func(t *testing.T) {
+			modelfile, err := ParseFile(strings.NewReader(c.input))
+			assert.ErrorIs(t, err, c.err)
+			if modelfile != nil {
+				assert.Equal(t, c.expected, modelfile.Commands)
+			}
+		})
+	}
+}
+
+func TestParseFileParametersMissingValue(t *testing.T) {
+	input := `
+FROM foo
+PARAMETER param1
+`
+
+	reader := strings.NewReader(input)
+
+	_, err := ParseFile(reader)
+	assert.ErrorIs(t, err, io.ErrUnexpectedEOF)
+}
+
+func TestParseFileBadCommand(t *testing.T) {
+	input := `
+FROM foo
+BADCOMMAND param1 value1
+`
+	_, err := ParseFile(strings.NewReader(input))
+	assert.ErrorIs(t, err, errInvalidCommand)
+
+}
+
+func TestParseFileMessages(t *testing.T) {
+	var cases = []struct {
+		input    string
+		expected []Command
+		err      error
+	}{
+		{
+			`
+FROM foo
+MESSAGE system You are a file parser. Always parse things.
+`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "message", Args: "system: You are a file parser. Always parse things."},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+MESSAGE system You are a file parser. Always parse things.`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "message", Args: "system: You are a file parser. Always parse things."},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+MESSAGE system You are a file parser. Always parse things.
+MESSAGE user Hey there!
+MESSAGE assistant Hello, I want to parse all the things!
+`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "message", Args: "system: You are a file parser. Always parse things."},
+				{Name: "message", Args: "user: Hey there!"},
+				{Name: "message", Args: "assistant: Hello, I want to parse all the things!"},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+MESSAGE system """
+You are a multiline file parser. Always parse things.
+"""
+			`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "message", Args: "system: \nYou are a multiline file parser. Always parse things.\n"},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+MESSAGE badguy I'm a bad guy!
+`,
+			nil,
+			errInvalidMessageRole,
+		},
+		{
+			`
+FROM foo
+MESSAGE system
+`,
+			nil,
+			io.ErrUnexpectedEOF,
+		},
+		{
+			`
+FROM foo
+MESSAGE system`,
+			nil,
+			io.ErrUnexpectedEOF,
+		},
+	}
+
+	for _, c := range cases {
+		t.Run("", func(t *testing.T) {
+			modelfile, err := ParseFile(strings.NewReader(c.input))
+			assert.ErrorIs(t, err, c.err)
+			if modelfile != nil {
+				assert.Equal(t, c.expected, modelfile.Commands)
+			}
+		})
+	}
+}
+
+func TestParseFileQuoted(t *testing.T) {
+	var cases = []struct {
+		multiline string
+		expected  []Command
+		err       error
+	}{
+		{
+			`
+FROM foo
+SYSTEM """
+This is a
+multiline system.
+"""
+			`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: "\nThis is a\nmultiline system.\n"},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+SYSTEM """
+This is a
+multiline system."""
+			`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: "\nThis is a\nmultiline system."},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+SYSTEM """This is a
+multiline system."""
+			`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: "This is a\nmultiline system."},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+SYSTEM """This is a multiline system."""
+			`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: "This is a multiline system."},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+SYSTEM """This is a multiline system.""
+			`,
+			nil,
+			io.ErrUnexpectedEOF,
+		},
+		{
+			`
+FROM foo
+SYSTEM "
+			`,
+			nil,
+			io.ErrUnexpectedEOF,
+		},
+		{
+			`
+FROM foo
+SYSTEM """
+This is a multiline system with "quotes".
+"""
+`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: "\nThis is a multiline system with \"quotes\".\n"},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+SYSTEM """"""
+`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: ""},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+SYSTEM ""
+`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: ""},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+SYSTEM "'"
+`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: "'"},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+SYSTEM """''"'""'""'"'''''""'""'"""
+`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "system", Args: `''"'""'""'"'''''""'""'`},
+			},
+			nil,
+		},
+		{
+			`
+FROM foo
+TEMPLATE """
+{{ .Prompt }}
+"""`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+				{Name: "template", Args: "\n{{ .Prompt }}\n"},
+			},
+			nil,
+		},
+	}
+
+	for _, c := range cases {
+		t.Run("", func(t *testing.T) {
+			modelfile, err := ParseFile(strings.NewReader(c.multiline))
+			assert.ErrorIs(t, err, c.err)
+			if modelfile != nil {
+				assert.Equal(t, c.expected, modelfile.Commands)
+			}
+		})
+	}
+}
+
+func TestParseFileParameters(t *testing.T) {
+	var cases = map[string]struct {
+		name, value string
+	}{
+		"numa true":                    {"numa", "true"},
+		"num_ctx 1":                    {"num_ctx", "1"},
+		"num_batch 1":                  {"num_batch", "1"},
+		"num_gqa 1":                    {"num_gqa", "1"},
+		"num_gpu 1":                    {"num_gpu", "1"},
+		"main_gpu 1":                   {"main_gpu", "1"},
+		"low_vram true":                {"low_vram", "true"},
+		"f16_kv true":                  {"f16_kv", "true"},
+		"logits_all true":              {"logits_all", "true"},
+		"vocab_only true":              {"vocab_only", "true"},
+		"use_mmap true":                {"use_mmap", "true"},
+		"use_mlock true":               {"use_mlock", "true"},
+		"num_thread 1":                 {"num_thread", "1"},
+		"num_keep 1":                   {"num_keep", "1"},
+		"seed 1":                       {"seed", "1"},
+		"num_predict 1":                {"num_predict", "1"},
+		"top_k 1":                      {"top_k", "1"},
+		"top_p 1.0":                    {"top_p", "1.0"},
+		"tfs_z 1.0":                    {"tfs_z", "1.0"},
+		"typical_p 1.0":                {"typical_p", "1.0"},
+		"repeat_last_n 1":              {"repeat_last_n", "1"},
+		"temperature 1.0":              {"temperature", "1.0"},
+		"repeat_penalty 1.0":           {"repeat_penalty", "1.0"},
+		"presence_penalty 1.0":         {"presence_penalty", "1.0"},
+		"frequency_penalty 1.0":        {"frequency_penalty", "1.0"},
+		"mirostat 1":                   {"mirostat", "1"},
+		"mirostat_tau 1.0":             {"mirostat_tau", "1.0"},
+		"mirostat_eta 1.0":             {"mirostat_eta", "1.0"},
+		"penalize_newline true":        {"penalize_newline", "true"},
+		"stop ### User:":               {"stop", "### User:"},
+		"stop ### User: ":              {"stop", "### User: "},
+		"stop \"### User:\"":           {"stop", "### User:"},
+		"stop \"### User: \"":          {"stop", "### User: "},
+		"stop \"\"\"### User:\"\"\"":   {"stop", "### User:"},
+		"stop \"\"\"### User:\n\"\"\"": {"stop", "### User:\n"},
+		"stop <|endoftext|>":           {"stop", "<|endoftext|>"},
+		"stop <|eot_id|>":              {"stop", "<|eot_id|>"},
+		"stop </s>":                    {"stop", "</s>"},
+	}
+
+	for k, v := range cases {
+		t.Run(k, func(t *testing.T) {
+			var b bytes.Buffer
+			fmt.Fprintln(&b, "FROM foo")
+			fmt.Fprintln(&b, "PARAMETER", k)
+			modelfile, err := ParseFile(&b)
+			assert.NoError(t, err)
+
+			assert.Equal(t, []Command{
+				{Name: "model", Args: "foo"},
+				{Name: v.name, Args: v.value},
+			}, modelfile.Commands)
+		})
+	}
+}
+
+func TestParseFileComments(t *testing.T) {
+	var cases = []struct {
+		input    string
+		expected []Command
+	}{
+		{
+			`
+# comment
+FROM foo
+	`,
+			[]Command{
+				{Name: "model", Args: "foo"},
+			},
+		},
+	}
+
+	for _, c := range cases {
+		t.Run("", func(t *testing.T) {
+			modelfile, err := ParseFile(strings.NewReader(c.input))
+			assert.NoError(t, err)
+			assert.Equal(t, c.expected, modelfile.Commands)
+		})
+	}
+}
+
+func TestParseFileFormatParseFile(t *testing.T) {
+	var cases = []string{
+		`
+FROM foo
+ADAPTER adapter1
+LICENSE MIT
+PARAMETER param1 value1
+PARAMETER param2 value2
+TEMPLATE template1
+MESSAGE system You are a file parser. Always parse things.
+MESSAGE user Hey there!
+MESSAGE assistant Hello, I want to parse all the things!
+`,
+		`
+FROM foo
+ADAPTER adapter1
+LICENSE MIT
+PARAMETER param1 value1
+PARAMETER param2 value2
+TEMPLATE template1
+MESSAGE system """
+You are a store greeter. Always responsed with "Hello!".
+"""
+MESSAGE user Hey there!
+MESSAGE assistant Hello, I want to parse all the things!
+`,
+		`
+FROM foo
+ADAPTER adapter1
+LICENSE """
+Very long and boring legal text.
+Blah blah blah.
+"Oh look, a quote!"
+"""
+
+PARAMETER param1 value1
+PARAMETER param2 value2
+TEMPLATE template1
+MESSAGE system """
+You are a store greeter. Always responsed with "Hello!".
+"""
+MESSAGE user Hey there!
+MESSAGE assistant Hello, I want to parse all the things!
+`,
+	}
+
+	for _, c := range cases {
+		t.Run("", func(t *testing.T) {
+			modelfile, err := ParseFile(strings.NewReader(c))
+			assert.NoError(t, err)
+
+			modelfile2, err := ParseFile(strings.NewReader(modelfile.String()))
+			assert.NoError(t, err)
+
+			assert.Equal(t, modelfile, modelfile2)
+		})
+	}
+
+}
Author	SHA1	Message	Date
Bruce MacDonald	d40497b9a2	Add support for IQ1_S, IQ3_S, IQ2_S, IQ4_XS. IQ4_NL Co-authored-by: ManniX-ITA <20623405+mann1x@users.noreply.github.com>	2024-05-03 14:51:07 -07:00
Michael Yang	828e4bf101	s/DisplayLongest/String/	2024-05-03 13:18:28 -07:00
Michael Yang	05105903d8	only quantize language models	2024-05-03 13:18:28 -07:00
Michael Yang	abf3b1fb34	no iterator	2024-05-03 13:18:28 -07:00
Michael Yang	82fcc0601d	rebase	2024-05-03 13:18:28 -07:00
Michael Yang	185a927210	comments	2024-05-03 13:18:28 -07:00
Michael Yang	096ea2c8c3	update tests	2024-05-03 13:18:28 -07:00
Michael Yang	06b31e2e24	quantize any fp16/fp32 model - FROM /path/to/{safetensors,pytorch} - FROM /path/to/fp{16,32}.bin - FROM model:fp{16,32}	2024-05-03 13:18:28 -07:00
Michael Yang	b7a87a22b6	Merge pull request #4059 from ollama/mxyng/parser-2 rename parser to model/file	2024-05-03 13:01:22 -07:00
Dr Nic Williams	e8aaea030e	Update 'llama2' -> 'llama3' in most places (#4116 ) * Update 'llama2' -> 'llama3' in most places --------- Co-authored-by: Patrick Devine <patrick@infrahq.com>	2024-05-03 15:25:04 -04:00
Daniel Hiltgen	267e25a750	Merge pull request #4129 from dhiltgen/unit_tests Soften timeouts on sched unit tests	2024-05-03 11:10:26 -07:00
Daniel Hiltgen	9a32c514cb	Soften timeouts on sched unit tests This gives us more headroom on the scheduler tests to tamp down some flakes.	2024-05-03 09:08:33 -07:00
Michael Yang	e9ae607ece	Merge pull request #3892 from ollama/mxyng/parser refactor modelfile parser	2024-05-02 17:04:47 -07:00
Michael Yang	93707fa3f2	Merge pull request #4108 from ollama/mxyng/lf fix line ending	2024-05-02 14:55:15 -07:00
Michael Yang	94c369095f	fix line ending replace CRLF with LF	2024-05-02 14:53:13 -07:00
Michael Yang	8acb233668	use strings.Builder	2024-05-01 10:01:09 -07:00
Michael Yang	119589fcb3	rename parser to model/file	2024-05-01 09:53:50 -07:00
Michael Yang	5ea844964e	cmd: import regexp	2024-05-01 09:53:45 -07:00
Michael Yang	bd8eed57fc	fix parser name	2024-05-01 09:52:54 -07:00
Michael Yang	9cf0f2e973	use parser.Format instead of templating modelfile	2024-05-01 09:52:54 -07:00
Michael Yang	176ad3aa6e	parser: add commands format	2024-05-01 09:52:54 -07:00
Michael Yang	4d08363580	comments	2024-05-01 09:52:54 -07:00
Michael Yang	8907bf51d2	fix multiline	2024-05-01 09:52:54 -07:00
Michael Yang	abe614c705	tests	2024-05-01 09:52:54 -07:00
Michael Yang	238715037d	linting	2024-05-01 09:52:54 -07:00
Michael Yang	c0a00f68ae	refactor modelfile parser	2024-05-01 09:52:54 -07:00