Commit Graph
5087 Commits
Author SHA1 Message Date
Parth SareenandGitHub 8a257ec00a docs: make integrations more discoverable (#14301)
* docs: add Pi integration page

* docs: flatten integration sidebar with expanded subheadings

* docs: add OpenClaw and Claude Code to quickstart
2026-02-17 13:27:25 -08:00
Parth SareenandGitHub 2f4de1acf7 cmd: ollama launch always show model picker (#14299) 2026-02-17 12:02:14 -08:00
Parth SareenandGitHub ec95c45f70 cmd/config: ollama launch cline CLI (#14294) 2026-02-17 11:37:53 -08:00
Patrick DevineandGitHub 3a88f7eb20 bugfix: add missing linear layer factory (#14289) 2026-02-16 17:22:20 -08:00
Patrick DevineandGitHub 0d5da826d4 bugfix: display the parameter count correctly in mlx for ollama show (#14285) 2026-02-16 13:03:34 -08:00
Patrick DevineandGitHub 9b795698b8 model: add llama3 architecture to mlxrunner (#14277) 2026-02-15 23:06:28 -08:00
Patrick DevineandGitHub 041fb77639 model: add gemma3 to the mlxrunner (#14276)
This change adds the gemma3 model to the mlxrunner and simplifies some of the quantization
code for loading weights.
2026-02-15 22:47:59 -08:00
Saumil ShahandGitHub 8224cce583 readme: update download link for macOS (#1) (#14271) 2026-02-15 15:25:15 -08:00
Patrick DevineandGitHub d18dcd7775 mlxrunner fixes (#14247)
* load glm4_moe_lite from the mlxrunner

* fix loading diffusion models

* remove log lines

* fix --imagegen flag
v0.16.2-rc0 v0.16.2
2026-02-13 22:30:42 -08:00
Parth SareenandGitHub 5f5ef20131 anthropic: enable websearch (#14246) 2026-02-13 19:20:46 -08:00
Parth SareenandGitHub f0a07a353b cmd/tui: fix powershell search (#14242) 2026-02-13 15:53:11 -08:00
948de6bbd2 add ability to disable cloud (#14221)
* add ability to disable cloud

Users can now easily opt-out of cloud inference and web search by
setting

```
"disable_ollama_cloud": true
```

in their `~/.ollama/server.json` settings file. After a setting update,
the server must be restarted.

Alternatively, setting the environment variable `OLLAMA_NO_CLOUD=1` will
also disable cloud features. While users previously were able to avoid
cloud models by not pulling or `ollama run`ing them, this gives them an
easy way to enforce that decision. Any attempt to run a cloud model when
cloud is disabled will fail.

The app's old "airplane mode" setting, which did a similar thing for
hiding cloud models within the app is now unified with this new cloud
disabled mode. That setting has been replaced with a "Cloud" toggle,
which behind the scenes edits `server.json` and then restarts the
server.

* gate cloud models across TUI and launch flows when cloud is disabled

Block cloud models from being selected, launched, or written to
integration configs when cloud mode is turned off:

- TUI main menu: open model picker instead of launching with a
  disabled cloud model
- cmd.go: add IsCloudModelDisabled checks for all Selection* paths
- LaunchCmd: filter cloud models from saved Editor configs before
  launch, fall through to picker if none remain
- Editor Run() methods (droid, opencode, openclaw): filter cloud
  models before calling Edit() and persist the cleaned list
- Export SaveIntegration, remove SaveIntegrationModel wrapper that
  was accumulating models instead of replacing them

* rename saveIntegration to SaveIntegration in config.go and tests

* cmd/config: add --model guarding and empty model list fixes

* Update docs/faq.mdx

Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>

* Update internal/cloud/policy.go

Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>

* Update internal/cloud/policy.go

Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>

* Update server/routes.go

Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>

* Revert "Update internal/cloud/policy.go"

This reverts commit 8bff8615f9.

Since this error shows up in other integrations, we want it to be
prefixed with Ollama

* rename cloud status

* more status renaming

* fix tests that weren't updated after rename

---------

Co-authored-by: ParthSareen <parth.sareen@ollama.com>
Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>
2026-02-12 15:47:00 -08:00
Parth SareenandGitHub 598b74d42c cmd/config: add minimax-m2.5 (#14223) v0.16.1 2026-02-12 14:29:50 -08:00
Jeffrey MorganandGitHub 935a48ed1a scripts: skip macOS symlink creation if already correct (#14142) 2026-02-12 12:44:42 -08:00
Daniel HiltgenandGitHub de39e24bf7 win: progress reporting on install download (#14219)
* win: progress reporting on install download

Downloading Ollama...
  [####################################    ] 91%  1106.6 / 1204.2 MB

* review comments
2026-02-12 12:06:56 -08:00
Eva HandGitHub 519b11eba1 site: update readme (#14217) 2026-02-12 12:14:13 -05:00
Eva HandGitHub 379fd64fa8 Revert "update README (#14213)" (#14215) 2026-02-12 12:06:00 -05:00
59c019a6fb x: configurable model load timeout (#14204)
Co-authored-by: rick <rick@frob.com.au>
2026-02-12 09:05:42 -08:00
Eva HandGitHub fad3bcccb2 update README (#14213) 2026-02-12 11:59:42 -05:00
Bruce MacDonaldandGitHub bd6697ad95 docs: update quickstart for tui (#14208) 2026-02-12 08:44:33 -08:00
SamareshSinghandGitHub f8dc7c9f54 docs: fix openapi schema for /api/ps and /api/tags endpoints (#14210) v0.16.0 v0.16.0-rc2 2026-02-11 17:37:40 -08:00
Patrick DevineandGitHub 4a3741129d bug: fix loading non-mlx models when ollama is built with mlx support (#14211)
This change fixes an issue where GGML based models (for either the Ollama runner or
the legacy llama.cpp runner) would try to load the mlx library. That would panic
and the model fails to start.
v0.16.0-rc1 v0.16.0-rc0
2026-02-11 14:48:33 -08:00
Parth SareenandGitHub 77ba9404ac cmd/tui: improve model picker UX (#14209) 2026-02-11 14:36:54 -08:00
Patrick DevineandGitHub 0aaf6119ec feature: add ctrl-g to allow users to use an editor to edit their prompt (#14197) 2026-02-11 13:04:41 -08:00
Parth SareenandGitHub f08427c138 cmd: TUI UX improvements (#14198) 2026-02-11 10:18:41 -08:00
MaternionandJesse Gross 2dbb000908 update context length format. 2026-02-10 17:06:05 -08:00
MaternionandJesse Gross c980e19995 Fix formatting of context length notes in documentation 2026-02-10 17:06:05 -08:00
MaternionandJesse Gross 6162374ca9 Update context-length.mdx 2026-02-10 17:06:05 -08:00
44bdd9a2ef Add MLX runner with GLM4-MoE-Lite model support (#14185)
This change adds a new MLX based runner which includes:

  * Method-based MLX bindings
  * Subprocess-based MLX runner (x/mlxrunner)
  * KV cache with tree management
  * A basic sampler

The GLM4-MoE-Lite model has been ported to use the new bindings.

---------

Co-authored-by: Michael Yang <git@mxy.ng>
2026-02-10 14:57:57 -08:00
MichaelandGitHub db493d6e5e docs: update broken links on FAQ and quick cleanup (#14194)
docs: update broken links on FAQ and quick cleanup
2026-02-10 16:52:20 -05:00
Bruce MacDonaldandGitHub 75695f16a5 docs: integration overview (#13831)
Group integrations into high-level types
2026-02-10 11:41:09 -08:00
Patrick DevineandGitHub a0407d07fa safetensors quantization for mlx (#14184)
This change includes:
  - changes to the safetensors metadata format
  - changes to the create command to properly create the blobs with the new format
  - changes to load the new format
  - fixes ollama show to properly show each tensor
2026-02-10 11:29:17 -08:00
Jeffrey MorganandGitHub 9ec733e527 cmd: make 'ollama login' and 'ollama logout' aliases for 'ollama signin' and 'ollama signout' respectively (#14144) 2026-02-09 19:12:42 -08:00
Parth SareenandGitHub 5ef04dab52 cmd: ollama launch pi (#14084) 2026-02-09 19:07:41 -08:00
Daniel HiltgenandGitHub aea316f1e9 win: add curl-style install script (#14178)
This adds a new powershell install script suitable for running via

  irm https://ollama.com/install.ps1 | iex

If you download the script and run '-?' it reports basic usage
information, as well as usage examples for common customization
options.  The script is signed as part of the release process
to ensure it can run on a typically configured Windows system.

This does not include doc updates - we can merge those after a release
ships to avoid user confusion.
2026-02-09 15:28:11 -08:00
Patrick DevineandGitHub 235ba3df5c cmd: ollama menu and launch improvements (#14038) 2026-02-09 11:30:16 -08:00
Jeffrey MorganandGitHub 099a0f18ef build: fix Dockerfile mlx directory (#14131) v0.15.6 2026-02-06 17:08:53 -08:00
Richard LyonsandJesse Gross fff696ee31 docs: increased RAM requirement for parallelism 2026-02-06 15:49:39 -08:00
Jeffrey MorganandGitHub 2e3ce6eab3 anthropic: do not count image tokens for now (#14127) 2026-02-06 15:33:18 -08:00
Parth SareenandGitHub 9e2003f88a cmd/config: offer to pull missing models instead of erroring (#14113) 2026-02-06 10:19:47 -08:00
Parth SareenandGitHub 42e1d49fbe cmd: fix context limits for droid and add qwen3-coder-next ctx (#14112) 2026-02-05 22:29:53 -08:00
Michael YangandGitHub 814630ca60 Revert "move tokenizers to separate package (#13825)" (#14111) 2026-02-05 20:49:08 -08:00
Parth SareenandGitHub 87cf187774 cmd: set claude code env vars on launch (#14109)
Set ANTHROPIC_DEFAULT_OPUS_MODEL, ANTHROPIC_DEFAULT_SONNET_MODEL,
ANTHROPIC_DEFAULT_HAIKU_MODEL, and CLAUDE_CODE_SUBAGENT_MODEL when
launching Claude Code so all model tiers route through Ollama.
2026-02-05 19:04:53 -08:00
Michael YangandGitHub 6ddd8862cd chore: move x/mlxrunner into x/imagegen (#14100) 2026-02-05 18:25:56 -08:00
Michael YangandGitHub f1373193dc move tokenizers to separate package (#13825) 2026-02-05 17:44:11 -08:00
Parth SareenandGitHub 8a4b77f9da cmd: set context limits for cloud models in opencode (#14107) v0.15.5 2026-02-05 16:36:46 -08:00
Parth SareenandGitHub 5f53fe7884 cmd: ollama launch improvements (#14099) v0.15.5-rc5 2026-02-05 15:08:17 -08:00
Bruce MacDonaldandGitHub 7ab4ca0e7f scripts: add macOS support to install.sh (#14060)
Allow installing Ollama on MacOS directly from the command line. This is in line with other CLI tools and results in a more streamlined experience when the user is looking to use the CLI specifically.
2026-02-05 14:59:01 -08:00
Jeffrey MorganandGitHub e36f389e82 scheduler: default parallel=1 for qwen3next/lfm (#14103) 2026-02-05 12:48:25 -08:00
Jesse Gross c61023f554 ollamarunner: Fix off by one error with numPredict
When numPredict is set, the user will receive one less token
than the requested limit. In addition, the stats will incorrectly
show the number of tokens returned as the limit. In cases where
numPredict is not set, the number of tokens is reported correctly.

This occurs because numPredict is checked when setting up the next
batch but hitting the limit will terminate the current batch as well.
Instead, is is better to check the limit as we actually predict them.
v0.15.5-rc4
2026-02-04 17:14:24 -08:00