errorsis

.github/workflows/release.yaml

@@ -31,7 +31,7 @@ jobs:
           security set-keychain-settings -lut 3600 build.keychain
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - name: Build Darwin
         env:
@@ -87,7 +87,7 @@ jobs:
           write-host "plugin installed"
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - run: go get ./...
       - run: |
@@ -141,7 +141,7 @@ jobs:
           write-host "plugin installed"
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - name: 'Install ROCm'
         run: |
@@ -218,7 +218,7 @@ jobs:
           write-host "plugin installed"
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - name: 'Install CUDA'
         run: |
@@ -306,7 +306,7 @@ jobs:
           write-host "plugin installed"
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - run: go get
       - uses: actions/download-artifact@v4

.github/workflows/test.yaml

@@ -63,7 +63,7 @@ jobs:
       - uses: actions/checkout@v4
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - run: go get ./...
       - run: |
@@ -163,7 +163,7 @@ jobs:
       - uses: actions/checkout@v4
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - name: 'Install ROCm'
         run: |
@@ -200,7 +200,7 @@ jobs:
       - uses: actions/checkout@v4
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - name: 'Install CUDA'
         run: |
@@ -255,7 +255,7 @@ jobs:
           submodules: recursive
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: false
       - run: |
           case ${{ matrix.arch }} in
@@ -297,7 +297,7 @@ jobs:
           submodules: recursive
       - uses: actions/setup-go@v5
         with:
-          go-version: "stable"
+          go-version-file: go.mod
           cache: true
       - run: |
           case ${{ matrix.arch }} in

Dockerfile

@@ -1,4 +1,4 @@
-ARG GOLANG_VERSION=1.22.5
+ARG GOLANG_VERSION=1.22.1
 ARG CMAKE_VERSION=3.22.1
 # this CUDA_VERSION corresponds with the one specified in docs/gpu.md
 ARG CUDA_VERSION=11.3.1

README.md

@@ -296,7 +296,6 @@ See the [API documentation](./docs/api.md) for all endpoints.
 - [Kerlig AI](https://www.kerlig.com/) (AI writing assistant for macOS)
 - [AI Studio](https://github.com/MindWorkAI/AI-Studio)
 - [Sidellama](https://github.com/gyopak/sidellama) (browser-based LLM client)
-- [LLMStack](https://github.com/trypromptly/LLMStack) (No-code multi-agent framework to build LLM agents and workflows)
 
 ### Terminal
 

auth/auth.go

@@ -3,49 +3,68 @@ package auth
 import (
 	"bytes"
 	"context"
+	"crypto/ed25519"
 	"crypto/rand"
 	"encoding/base64"
+	"encoding/pem"
+	"errors"
 	"fmt"
 	"io"
 	"log/slog"
 	"os"
 	"path/filepath"
-	"strings"
 
 	"golang.org/x/crypto/ssh"
 )
 
 const defaultPrivateKey = "id_ed25519"
 
-func keyPath() (string, error) {
+func privateKey() (ssh.Signer, error) {
 	home, err := os.UserHomeDir()
 	if err != nil {
-		return "", err
+		return nil, err
 	}
 
-	return filepath.Join(home, ".ollama", defaultPrivateKey), nil
+	keyPath := filepath.Join(home, ".ollama", defaultPrivateKey)
+	privateKeyFile, err := os.ReadFile(keyPath)
+	if errors.Is(err, os.ErrNotExist) {
+		err := initializeKeypair()
+		if err != nil {
+			return nil, err
+		}
+
+		return privateKey()
+	} else if err != nil {
+		slog.Info(fmt.Sprintf("Failed to load private key: %v", err))
+		return nil, err
+	}
+
+	return ssh.ParsePrivateKey(privateKeyFile)
 }
 
-func GetPublicKey() (string, error) {
-	keyPath, err := keyPath()
+func GetPublicKey() (ssh.PublicKey, error) {
+	// try to read pubkey first
+	home, err := os.UserHomeDir()
 	if err != nil {
-		return "", err
+		return nil, err
 	}
 
-	privateKeyFile, err := os.ReadFile(keyPath)
-	if err != nil {
-		slog.Info(fmt.Sprintf("Failed to load private key: %v", err))
-		return "", err
+	pubkeyPath := filepath.Join(home, ".ollama", defaultPrivateKey+".pub")
+	pubKeyFile, err := os.ReadFile(pubkeyPath)
+	if errors.Is(err, os.ErrNotExist) {
+		// try from privateKey
+		privateKey, err := privateKey()
+		if err != nil {
+			return nil, fmt.Errorf("failed to read public key: %w", err)
+		}
+
+		return privateKey.PublicKey(), nil
+	} else if err != nil {
+		return nil, fmt.Errorf("failed to read public key: %w", err)
 	}
 
-	privateKey, err := ssh.ParsePrivateKey(privateKeyFile)
-	if err != nil {
-		return "", err
-	}
-
-	publicKey := ssh.MarshalAuthorizedKey(privateKey.PublicKey())
-
-	return strings.TrimSpace(string(publicKey)), nil
+	pubKey, _, _, _, err := ssh.ParseAuthorizedKey(pubKeyFile)
+	return pubKey, err
 }
 
 func NewNonce(r io.Reader, length int) (string, error) {
@@ -58,25 +77,20 @@ func NewNonce(r io.Reader, length int) (string, error) {
 }
 
 func Sign(ctx context.Context, bts []byte) (string, error) {
-	keyPath, err := keyPath()
-	if err != nil {
-		return "", err
-	}
-
-	privateKeyFile, err := os.ReadFile(keyPath)
-	if err != nil {
-		slog.Info(fmt.Sprintf("Failed to load private key: %v", err))
-		return "", err
-	}
-
-	privateKey, err := ssh.ParsePrivateKey(privateKeyFile)
+	privateKey, err := privateKey()
 	if err != nil {
 		return "", err
 	}
 
 	// get the pubkey, but remove the type
-	publicKey := ssh.MarshalAuthorizedKey(privateKey.PublicKey())
-	parts := bytes.Split(publicKey, []byte(" "))
+	publicKey, err := GetPublicKey()
+	if err != nil {
+		return "", err
+	}
+
+	publicKeyBytes := ssh.MarshalAuthorizedKey(publicKey)
+
+	parts := bytes.Split(publicKeyBytes, []byte(" "))
 	if len(parts) < 2 {
 		return "", fmt.Errorf("malformed public key")
 	}
@@ -89,3 +103,49 @@ func Sign(ctx context.Context, bts []byte) (string, error) {
 	// signature is <pubkey>:<signature>
 	return fmt.Sprintf("%s:%s", bytes.TrimSpace(parts[1]), base64.StdEncoding.EncodeToString(signedData.Blob)), nil
 }
+
+func initializeKeypair() error {
+	home, err := os.UserHomeDir()
+	if err != nil {
+		return err
+	}
+
+	privKeyPath := filepath.Join(home, ".ollama", "id_ed25519")
+	pubKeyPath := filepath.Join(home, ".ollama", "id_ed25519.pub")
+
+	_, err = os.Stat(privKeyPath)
+	if errors.Is(err, os.ErrNotExist) {
+		fmt.Printf("Couldn't find '%s'. Generating new private key.\n", privKeyPath)
+		cryptoPublicKey, cryptoPrivateKey, err := ed25519.GenerateKey(rand.Reader)
+		if err != nil {
+			return err
+		}
+
+		privateKeyBytes, err := ssh.MarshalPrivateKey(cryptoPrivateKey, "")
+		if err != nil {
+			return err
+		}
+
+		if err := os.MkdirAll(filepath.Dir(privKeyPath), 0o755); err != nil {
+			return fmt.Errorf("could not create directory %w", err)
+		}
+
+		if err := os.WriteFile(privKeyPath, pem.EncodeToMemory(privateKeyBytes), 0o600); err != nil {
+			return err
+		}
+
+		sshPublicKey, err := ssh.NewPublicKey(cryptoPublicKey)
+		if err != nil {
+			return err
+		}
+
+		publicKeyBytes := ssh.MarshalAuthorizedKey(sshPublicKey)
+
+		if err := os.WriteFile(pubKeyPath, publicKeyBytes, 0o644); err != nil {
+			return err
+		}
+
+		fmt.Printf("Your new public key is: \n\n%s\n", publicKeyBytes)
+	}
+	return nil
+}

cmd/cmd.go

@@ -4,10 +4,7 @@ import (
 	"archive/zip"
 	"bytes"
 	"context"
-	"crypto/ed25519"
-	"crypto/rand"
 	"crypto/sha256"
-	"encoding/pem"
 	"errors"
 	"fmt"
 	"io"
@@ -379,11 +376,12 @@ func errFromUnknownKey(unknownKeyErr error) error {
 	if len(matches) > 0 {
 		serverPubKey := matches[0]
 
-		localPubKey, err := auth.GetPublicKey()
+		publicKey, err := auth.GetPublicKey()
 		if err != nil {
 			return unknownKeyErr
 		}
 
+		localPubKey := strings.TrimSpace(string(ssh.MarshalAuthorizedKey(publicKey)))
 		if runtime.GOOS == "linux" && serverPubKey != localPubKey {
 			// try the ollama service public key
 			svcPubKey, err := os.ReadFile("/usr/share/ollama/.ollama/id_ed25519.pub")
@@ -1072,7 +1070,7 @@ func generate(cmd *cobra.Command, opts runOptions) error {
 }
 
 func RunServer(cmd *cobra.Command, _ []string) error {
-	if err := initializeKeypair(); err != nil {
+	if _, err := auth.GetPublicKey(); err != nil {
 		return err
 	}
 
@@ -1089,52 +1087,6 @@ func RunServer(cmd *cobra.Command, _ []string) error {
 	return err
 }
 
-func initializeKeypair() error {
-	home, err := os.UserHomeDir()
-	if err != nil {
-		return err
-	}
-
-	privKeyPath := filepath.Join(home, ".ollama", "id_ed25519")
-	pubKeyPath := filepath.Join(home, ".ollama", "id_ed25519.pub")
-
-	_, err = os.Stat(privKeyPath)
-	if os.IsNotExist(err) {
-		fmt.Printf("Couldn't find '%s'. Generating new private key.\n", privKeyPath)
-		cryptoPublicKey, cryptoPrivateKey, err := ed25519.GenerateKey(rand.Reader)
-		if err != nil {
-			return err
-		}
-
-		privateKeyBytes, err := ssh.MarshalPrivateKey(cryptoPrivateKey, "")
-		if err != nil {
-			return err
-		}
-
-		if err := os.MkdirAll(filepath.Dir(privKeyPath), 0o755); err != nil {
-			return fmt.Errorf("could not create directory %w", err)
-		}
-
-		if err := os.WriteFile(privKeyPath, pem.EncodeToMemory(privateKeyBytes), 0o600); err != nil {
-			return err
-		}
-
-		sshPublicKey, err := ssh.NewPublicKey(cryptoPublicKey)
-		if err != nil {
-			return err
-		}
-
-		publicKeyBytes := ssh.MarshalAuthorizedKey(sshPublicKey)
-
-		if err := os.WriteFile(pubKeyPath, publicKeyBytes, 0o644); err != nil {
-			return err
-		}
-
-		fmt.Printf("Your new public key is: \n\n%s\n", publicKeyBytes)
-	}
-	return nil
-}
-
 func checkServerHeartbeat(cmd *cobra.Command, _ []string) error {
 	client, err := api.ClientFromEnvironment()
 	if err != nil {

convert/mistral.go

@@ -71,11 +71,6 @@ func (m *MistralModel) WriteGGUF(ws io.WriteSeeker) error {
 		"tokenizer.ggml.unknown_token_id": uint32(0),
 	}
 
-	if m.Params.HeadDimension > 0 {
-		kv["llama.attention.key_length"] = uint32(m.Params.HeadDimension)
-		kv["llama.attention.value_length"] = uint32(m.Params.HeadDimension)
-	}
-
 	return llm.NewGGUFV3(m.Params.ByteOrder).Encode(ws, kv, m.Tensors)
 }
 

docs/api.md

@@ -1026,7 +1026,7 @@ If `stream` is set to `false`, then the response is a single JSON object:
 ## Generate Embeddings
 
 ```shell
-POST /api/embed
+POST /api/embeddings
 ```
 
 Generate embeddings from a model
@@ -1034,11 +1034,10 @@ Generate embeddings from a model
 ### Parameters
 
 - `model`: name of model to generate embeddings from
-- `input`: text or list of text to generate embeddings for
+- `prompt`: text to generate embeddings for
 
 Advanced parameters:
 
-- `truncate`: truncates the end of each input to fit within context length. Returns error if `false` and context length is exceeded. Defaults to `true`
 - `options`: additional model parameters listed in the documentation for the [Modelfile](./modelfile.md#valid-parameters-and-values) such as `temperature`
 - `keep_alive`: controls how long the model will stay loaded into memory following the request (default: `5m`)
 
@@ -1047,9 +1046,9 @@ Advanced parameters:
 #### Request
 
 ```shell
-curl http://localhost:11434/api/embed -d '{
+curl http://localhost:11434/api/embeddings -d '{
   "model": "all-minilm",
-  "input": "Why is the sky blue?"
+  "prompt": "Here is an article about llamas..."
 }'
 ```
 
@@ -1057,35 +1056,10 @@ curl http://localhost:11434/api/embed -d '{
 
 ```json
 {
-  "model": "all-minilm",
-  "embeddings": [[
-    0.010071029, -0.0017594862, 0.05007221, 0.04692972, 0.054916814,
-    0.008599704, 0.105441414, -0.025878139, 0.12958129, 0.031952348
-  ]]
-}
-```
-
-#### Request (Multiple input)
-
-```shell
-curl http://localhost:11434/api/embed -d '{
-  "model": "all-minilm",
-  "input": ["Why is the sky blue?", "Why is the grass green?"]
-}'
-```
-
-#### Response
-
-```json
-{
-  "model": "all-minilm",
-  "embeddings": [[
-    0.010071029, -0.0017594862, 0.05007221, 0.04692972, 0.054916814,
-    0.008599704, 0.105441414, -0.025878139, 0.12958129, 0.031952348
-  ],[
-    -0.0098027075, 0.06042469, 0.025257962, -0.006364387, 0.07272725,
-    0.017194884, 0.09032035, -0.051705178, 0.09951512, 0.09072481
-  ]]
+  "embedding": [
+    0.5670403838157654, 0.009260174818336964, 0.23178744316101074, -0.2916173040866852, -0.8924556970596313,
+    0.8785552978515625, -0.34576427936553955, 0.5742510557174683, -0.04222835972905159, -0.137906014919281
+  ]
 }
 ```
 
@@ -1132,45 +1106,3 @@ A single JSON object will be returned.
   ]
 }
 ```
-
-## Generate Embedding
-
-> Note: this endpoint has been superseded by `/api/embed`
-
-```shell
-POST /api/embeddings
-```
-
-Generate embeddings from a model
-
-### Parameters
-
-- `model`: name of model to generate embeddings from
-- `prompt`: text to generate embeddings for
-
-Advanced parameters:
-
-- `options`: additional model parameters listed in the documentation for the [Modelfile](./modelfile.md#valid-parameters-and-values) such as `temperature`
-- `keep_alive`: controls how long the model will stay loaded into memory following the request (default: `5m`)
-
-### Examples
-
-#### Request
-
-```shell
-curl http://localhost:11434/api/embeddings -d '{
-  "model": "all-minilm",
-  "prompt": "Here is an article about llamas..."
-}'
-```
-
-#### Response
-
-```json
-{
-  "embedding": [
-    0.5670403838157654, 0.009260174818336964, 0.23178744316101074, -0.2916173040866852, -0.8924556970596313,
-    0.8785552978515625, -0.34576427936553955, 0.5742510557174683, -0.04222835972905159, -0.137906014919281
-  ]
-}
-```

integration/embed_test.go

@@ -4,45 +4,12 @@ package integration
 
 import (
 	"context"
-	"math"
 	"testing"
 	"time"
 
 	"github.com/ollama/ollama/api"
 )
 
-func floatsEqual32(a, b float32) bool {
-	return math.Abs(float64(a-b)) <= 1e-4
-}
-
-func floatsEqual64(a, b float64) bool {
-	return math.Abs(a-b) <= 1e-4
-}
-
-func TestAllMiniLMEmbeddings(t *testing.T) {
-	ctx, cancel := context.WithTimeout(context.Background(), 2*time.Minute)
-	defer cancel()
-
-	req := api.EmbeddingRequest{
-		Model:  "all-minilm",
-		Prompt: "why is the sky blue?",
-	}
-
-	res, err := embeddingTestHelper(ctx, t, req)
-
-	if err != nil {
-		t.Fatalf("error: %v", err)
-	}
-
-	if len(res.Embedding) != 384 {
-		t.Fatalf("expected 384 floats, got %d", len(res.Embedding))
-	}
-
-	if !floatsEqual64(res.Embedding[0], 0.06642947345972061) {
-		t.Fatalf("expected 0.06642947345972061, got %.16f", res.Embedding[0])
-	}
-}
-
 func TestAllMiniLMEmbed(t *testing.T) {
 	ctx, cancel := context.WithTimeout(context.Background(), 2*time.Minute)
 	defer cancel()
@@ -66,8 +33,8 @@ func TestAllMiniLMEmbed(t *testing.T) {
 		t.Fatalf("expected 384 floats, got %d", len(res.Embeddings[0]))
 	}
 
-	if !floatsEqual32(res.Embeddings[0][0], 0.010071031) {
-		t.Fatalf("expected 0.010071031, got %.8f", res.Embeddings[0][0])
+	if res.Embeddings[0][0] != 0.010071031 {
+		t.Fatalf("expected 0.010071031, got %f", res.Embeddings[0][0])
 	}
 }
 
@@ -94,12 +61,12 @@ func TestAllMiniLMBatchEmbed(t *testing.T) {
 		t.Fatalf("expected 384 floats, got %d", len(res.Embeddings[0]))
 	}
 
-	if !floatsEqual32(res.Embeddings[0][0], 0.010071031) || !floatsEqual32(res.Embeddings[1][0], -0.009802706) {
-		t.Fatalf("expected 0.010071031 and -0.009802706, got %.8f and %.8f", res.Embeddings[0][0], res.Embeddings[1][0])
+	if res.Embeddings[0][0] != 0.010071031 || res.Embeddings[1][0] != -0.009802706 {
+		t.Fatalf("expected 0.010071031 and -0.009802706, got %f and %f", res.Embeddings[0][0], res.Embeddings[1][0])
 	}
 }
 
-func TestAllMiniLMEmbedTruncate(t *testing.T) {
+func TestAllMiniLmEmbedTruncate(t *testing.T) {
 	ctx, cancel := context.WithTimeout(context.Background(), 2*time.Minute)
 	defer cancel()
 
@@ -168,22 +135,6 @@ func TestAllMiniLMEmbedTruncate(t *testing.T) {
 	}
 }
 
-func embeddingTestHelper(ctx context.Context, t *testing.T, req api.EmbeddingRequest) (*api.EmbeddingResponse, error) {
-	client, _, cleanup := InitServerConnection(ctx, t)
-	defer cleanup()
-	if err := PullIfMissing(ctx, client, req.Model); err != nil {
-		t.Fatalf("failed to pull model %s: %v", req.Model, err)
-	}
-
-	response, err := client.Embeddings(ctx, &req)
-
-	if err != nil {
-		return nil, err
-	}
-
-	return response, nil
-}
-
 func embedTestHelper(ctx context.Context, t *testing.T, req api.EmbedRequest) (*api.EmbedResponse, error) {
 	client, _, cleanup := InitServerConnection(ctx, t)
 	defer cleanup()

server/images.go

@@ -32,6 +32,7 @@ import (
 	"github.com/ollama/ollama/types/errtypes"
 	"github.com/ollama/ollama/types/model"
 	"github.com/ollama/ollama/version"
+	"golang.org/x/crypto/ssh"
 )
 
 var (
@@ -1088,11 +1089,12 @@ func makeRequestWithRetry(ctx context.Context, method string, requestURL *url.UR
 	if anonymous {
 		// no user is associated with the public key, and the request requires non-anonymous access
 		pubKey, nestedErr := auth.GetPublicKey()
+		localPubKey := strings.TrimSpace(string(ssh.MarshalAuthorizedKey(pubKey)))
 		if nestedErr != nil {
 			slog.Error(fmt.Sprintf("couldn't get public key: %v", nestedErr))
 			return nil, errUnauthorized
 		}
-		return nil, &errtypes.UnknownOllamaKey{Key: pubKey}
+		return nil, &errtypes.UnknownOllamaKey{Key: localPubKey}
 	}
 	// user is associated with the public key, but is not authorized to make the request
 	return nil, errUnauthorized

server/routes.go

@@ -609,9 +609,10 @@ func (s *Server) CreateModelHandler(c *gin.Context) {
 		defer cancel()
 
 		quantization := cmp.Or(r.Quantize, r.Quantization)
-		if err := CreateModel(ctx, name, filepath.Dir(r.Path), strings.ToUpper(quantization), f, fn); errors.Is(err, errBadTemplate) {
-			ch <- gin.H{"error": err.Error(), "status": http.StatusBadRequest}
-		} else if err != nil {
+		if err := CreateModel(ctx, name, filepath.Dir(r.Path), strings.ToUpper(quantization), f, fn); err != nil {
+			if errors.Is(err, errBadTemplate) {
+				ch <- gin.H{"error": err.Error(), "status": http.StatusBadRequest}
+			}
 			ch <- gin.H{"error": err.Error()}
 		}
 	}()