ParthSareen
a2a73ce5e0
wip!
2025-01-23 20:21:50 -08:00
ParthSareen
6ba557f25b
checkpoint
2025-01-23 09:46:14 -08:00
ParthSareen
a7c8cc06da
json checkpoint
2025-01-21 17:48:12 -08:00
Blake Mizerany
089bbb537d
grammar: introduce new grammar package
...
This package provides a way to convert JSON schemas to equivalent EBNF.
It is intended to be a replacement to llama.cpp's schema_to_grammar.
This is still an early version and does not yet support all JSON schema
features. The to-do list includes:
- minumum/maximum constraints on integer types
- minLength/maxLength constraints on string types
- defs and refs
2025-01-20 14:40:50 -08:00
ParthSareen
7cd9fbbbb1
improve temperature sampler
2025-01-20 13:53:07 -08:00
ParthSareen
b91487f289
Working json sampler
2025-01-15 14:08:54 -08:00
ParthSareen
5b19d4941a
addressing comments + cleanup
2025-01-14 16:23:48 -08:00
ParthSareen
5e73f24e16
sampling package
2025-01-14 16:23:38 -08:00
Michael Yang
4aac178cac
update ci
2025-01-14 14:25:56 -08:00
Michael Yang
800e21d060
sort devices by type
2025-01-14 14:25:56 -08:00
Michael Yang
5827999e9e
next ollama runner
...
implement llama and mllama model architectures in go using ggml (through
cgo)
2025-01-14 14:25:56 -08:00
Jeffrey Morgan
61676fb506
llama: move grammar tests to llama_test.go ( #8411 )
2025-01-14 12:55:45 -08:00
Bruce MacDonald
f6f3713001
convert: qwen2 from safetensors ( #8408 )
...
Add native support for converting Qwen2 family models (including Qwen2.5)
from safetensors to gguf format so we can run it.
2025-01-14 10:34:37 -08:00
Steve Berdy
a30f347201
readme: add LangChain for .NET to community integrations ( #8352 )
2025-01-14 09:37:35 -08:00
Jeffrey Morgan
74ea4fb604
remove .prettierrc.json ( #8413 )
2025-01-14 09:30:34 -08:00
Jeffrey Morgan
6982e9cc96
readme: remove link to missing page
2025-01-13 18:56:31 -08:00
Patrick Devine
ab39872cb4
add new create api doc ( #8388 )
2025-01-13 17:30:24 -08:00
Parth Sareen
84a2314463
examples: remove codified examples ( #8267 )
2025-01-13 11:26:22 -08:00
Jeffrey Morgan
17fcdea698
readme: move discord link
2025-01-12 22:45:47 -08:00
Patrick Devine
32bd37adf8
make the modelfile path relative for ollama create ( #8380 )
2025-01-10 16:14:08 -08:00
Michael Yang
9446c2c902
Merge pull request #8196 from ollama/mxyng/gods-v2
...
chore: upgrade to gods v2
2025-01-10 13:50:11 -08:00
Jeffrey Morgan
9aa141d023
readme: remove discord badge image for now
2025-01-09 22:02:18 -08:00
Patrick Devine
8bccae4f92
show a more descriptive error in the client if it is newer than the server ( #8351 )
2025-01-09 10:12:30 -08:00
isamu arimoto
6ae2adc1af
openai: accept additional headers to fix CORS errors ( #8343 )
2025-01-08 11:28:11 -08:00
Jeffrey Morgan
1deafd8254
llama: update vendored code to commit 46e3556 ( #8308 )
2025-01-08 11:22:01 -08:00
Michael
57f038ec7b
readme: add phi4 model ( #8350 )
2025-01-08 11:21:39 -08:00
frob
cdf3a181dc
Add CUSTOM_CPU_FLAGS to Dockerfile. ( #8284 )
...
* Add CUSTOM_CPU_FLAGS.
* fix golangci-lint error.
---------
Co-authored-by: Richard Lyons <rick@frob.com.au >
2025-01-06 09:17:19 -08:00
Ubaldo Porcheddu
3919f4ba3d
llama: fix runner api example url in README.md ( #8307 )
2025-01-04 15:45:16 -08:00
Bruce MacDonald
2d33c4e97d
discover: remove leading new-line for linter
2025-01-03 12:03:58 -08:00
Bruce MacDonald
29a8975c66
api: remove unused create fields
...
These fields are deprecated, but specifying them will not do anything. Removing them as the other deprecated fields will still work, but these do not, so they dont match our existing pattern.
2025-01-03 12:03:58 -08:00
Patrick Devine
86a622cbdc
Update the /api/create endpoint to use JSON ( #7935 )
...
Replaces `POST /api/create` to use JSON instead of a Modelfile.
This is a breaking change.
2024-12-31 18:02:30 -08:00
Jeffrey Morgan
459d822b51
readme: link header to ollama.com
2024-12-29 17:36:07 -05:00
Simon Schampijer
844899440a
examples: updated deprecated imports ( #3602 )
2024-12-29 14:36:25 -05:00
Anas Khan
103db4216d
docs: add /api/version endpoint documentation ( #8082 )
...
Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com >
2024-12-29 14:33:44 -05:00
Jeffrey Morgan
6daddcde01
readme: update import header
2024-12-29 14:12:23 -05:00
Emilien Lancelot
07f7e69b36
readme: add Yacana multi-agent framework to community integrations ( #7259 )
2024-12-28 15:05:57 -05:00
CIIDMike
b68e8e5727
docs: add syntax highlighting on Go template code blocks ( #8215 )
2024-12-27 13:17:49 -05:00
Adarsh Mishra
369fb529e2
readme: add TextLLaMA to community integrations
2024-12-27 13:16:06 -05:00
Jared Donnell
023e4bca14
readme: add neollama to terminal section of community integrations ( #8242 )
2024-12-25 17:16:11 -05:00
aritra saha
51af455f62
readme: add alpaca client application to community integrations ( #8227 )
2024-12-24 23:05:35 -05:00
Emanuil Rusev
ffe3549064
readme: add IntelliBar to community integrations ( #7950 )
2024-12-23 12:04:18 -05:00
湛露先生
928de9050e
server: reuse InvalidModelNameErrMsg type ( #8163 )
2024-12-23 10:38:34 -05:00
ItzCrazyKns
36aea6154a
readme: add Perplexica to community-integrations ( #8198 )
2024-12-22 20:04:01 -05:00
Patrick Devine
dd352ab27f
fix crash bug with /save when quotes are used ( #8208 )
2024-12-21 22:31:37 -08:00
Michael Yang
cb40d60469
chore: upgrade to gods v2
...
gods v2 uses go generics rather than interfaces which simplifies the
code considerably
2024-12-21 00:05:16 -08:00
Patrick Devine
d8bab8ea44
remove tutorials.md which pointed to removed tutorials ( #8189 )
2024-12-20 14:04:20 -08:00
Squishedmac
9ab62eb96f
update golang.org/x dependencies ( #8172 )
2024-12-20 09:29:30 -08:00
Parth Sareen
290cf2040a
llama: test key order preservation in schema_to_grammar ( #8078 )
...
This change adds a test to catch a regression in schema_to_grammar where
the order of keys in the JSON schema is not preserved in the generated
grammar, which is critical for step-by-step reasoning.
2024-12-18 19:44:50 -08:00
Jeffrey Morgan
a72f2dce45
scripts: sign renamed macOS binary ( #8131 )
2024-12-17 18:03:49 -08:00
Jesse Gross
08a832b482
llama: Ensure KV cache is fully defragmented.
...
Sometimes the KV cache requires defragmentation even without
triggering the threshold heuristic. In this case, decoding
will not being able to find a KV cache slot. This is particularly
difficult for the caller to handle if it happens in between
ubatches. To avoid this, we should immediately trigger a defrag.
In addition, a heavily fragmented cache can require more than
max_moves to defragment. Currently, we stop when we hit the limit
but this can leave a cache that still does not have adequate space
even after defragmentation is triggered. Instead, we should do
multiple batches of processing until everything is complete.
Fixes #7949
2024-12-17 14:01:19 -08:00
