From f10d4bfbcceff2e9bd256e58fc5a6bf4be1d0a6a Mon Sep 17 00:00:00 2001 From: The Librarian Date: Sat, 12 Sep 2026 04:00:05 +0000 Subject: [PATCH] Sync ollama docs from 53fed261 on 2026-09-12 --- .SYNC_INFO.md | 4 +- docs/api.md | 3 +- docs/api/usage.mdx | 6 ++- docs/capabilities/vision.mdx | 2 +- docs/docs.json | 2 + docs/images/launch-icons/deepseek-harness.svg | 4 ++ docs/integrations/claude-desktop.mdx | 35 +++++++++++-- docs/integrations/codex.mdx | 12 +++++ docs/integrations/deepseek-harness.mdx | 50 +++++++++++++++++++ docs/integrations/index.mdx | 10 +++- docs/modelfile.mdx | 2 +- docs/openapi.yaml | 15 ++++-- 12 files changed, 131 insertions(+), 14 deletions(-) create mode 100644 docs/images/launch-icons/deepseek-harness.svg create mode 100644 docs/integrations/deepseek-harness.mdx diff --git a/.SYNC_INFO.md b/.SYNC_INFO.md index 907214a..669f9e9 100644 --- a/.SYNC_INFO.md +++ b/.SYNC_INFO.md @@ -4,8 +4,8 @@ This is a mirror of the Ollama repository. **Synced from:** https://github.com/ollama/ollama.git **Branch:** main -**Commit:** 948f69330acf96a2310f1b53fdfc211731a386d8 -**Sync Date:** 2026-08-12 +**Commit:** 53fed26112817f7c55f664efb9e3f65f06cab7db +**Sync Date:** 2026-09-12 **Content:** Paths: docs --- diff --git a/docs/api.md b/docs/api.md index 95cac42..87ac17d 100644 --- a/docs/api.md +++ b/docs/api.md @@ -100,7 +100,8 @@ The final response in the stream also includes additional data about the generat - `total_duration`: time spent generating the response - `load_duration`: time spent in nanoseconds loading the model - `prompt_eval_count`: number of tokens in the prompt -- `prompt_eval_duration`: time spent in nanoseconds evaluating the prompt +- `prompt_eval_cached_count`: number of prompt tokens read from the cache +- `prompt_eval_duration`: time spent in nanoseconds evaluating uncached prompt tokens - `eval_count`: number of tokens in the response - `eval_duration`: time in nanoseconds spent generating the response - `context`: an encoding of the conversation used in this response, this can be sent in the next request to keep a conversational memory diff --git a/docs/api/usage.mdx b/docs/api/usage.mdx index 107bb0c..e9606a4 100644 --- a/docs/api/usage.mdx +++ b/docs/api/usage.mdx @@ -6,8 +6,9 @@ Ollama's API responses include metrics that can be used for measuring performanc * `total_duration`: How long the response took to generate * `load_duration`: How long the model took to load -* `prompt_eval_count`: How many input tokens were processed -* `prompt_eval_duration`: How long it took to evaluate the prompt +* `prompt_eval_count`: How many input tokens were in the prompt +* `prompt_eval_cached_count`: How many prompt tokens were read from the cache +* `prompt_eval_duration`: How long it took to evaluate the uncached prompt tokens * `eval_count`: How many output tokens were processes * `eval_duration`: How long it took to generate the output tokens @@ -27,6 +28,7 @@ For endpoints that return usage metrics, the response body will include the usag "total_duration": 174560334, "load_duration": 101397084, "prompt_eval_count": 11, + "prompt_eval_cached_count": 8, "prompt_eval_duration": 13074791, "eval_count": 18, "eval_duration": 52479709 diff --git a/docs/capabilities/vision.mdx b/docs/capabilities/vision.mdx index 1ecd50c..a8e77a4 100644 --- a/docs/capabilities/vision.mdx +++ b/docs/capabilities/vision.mdx @@ -7,7 +7,7 @@ Vision models accept images alongside text so the model can describe, classify, ## Quick start ```shell -ollama run gemma4 ./image.png whats in this image? +ollama run gemma4 ./image.png what is in this image? ``` diff --git a/docs/docs.json b/docs/docs.json index 0d67393..61d0753 100644 --- a/docs/docs.json +++ b/docs/docs.json @@ -157,6 +157,7 @@ "group": "Assistants", "expanded": true, "pages": [ + "/integrations/claude-desktop", "/integrations/openclaw", "/integrations/hermes", "/integrations/hermes-desktop" @@ -168,6 +169,7 @@ "pages": [ "/integrations/claude-code", "/integrations/opencode", + "/integrations/deepseek-harness", "/integrations/cline-cli", "/integrations/codex-app", "/integrations/codex", diff --git a/docs/images/launch-icons/deepseek-harness.svg b/docs/images/launch-icons/deepseek-harness.svg new file mode 100644 index 0000000..4f770e5 --- /dev/null +++ b/docs/images/launch-icons/deepseek-harness.svg @@ -0,0 +1,4 @@ + + + + diff --git a/docs/integrations/claude-desktop.mdx b/docs/integrations/claude-desktop.mdx index 66e1095..2792bff 100644 --- a/docs/integrations/claude-desktop.mdx +++ b/docs/integrations/claude-desktop.mdx @@ -1,13 +1,42 @@ --- title: Claude Desktop +description: Use Ollama models in Claude on macOS. --- -Claude Desktop is no longer supported by `ollama launch`. -Existing installations can be restored to the usual Claude profile: +## Prerequisites + +- [Ollama](https://ollama.com/download/). Windows support coming soon. +- To use a local model, [download it](/quickstart#3-start-a-chat) before setup. +- To use a cloud model, sign in to Ollama and enable cloud models. Some models require a [paid plan](https://ollama.com/pricing). + +If Claude is not installed, Ollama offers to download it during setup. + +## Set up Claude + +1. Open Ollama and select **Apps**. +2. Toggle Claude to On. +3. If Claude is not installed, Ollama will download and finish the installation. + +## Switch models + +Open Ollama **Settings**. Under **Apps**, choose a model you want to use, then select **Restart Claude**. You can assign the same Ollama model to more than one option. + +## Supported features + +- **Subagents** - Split larger tasks across agents +- **Web search** - Supported by default through [Ollama's web search](/capabilities/web-search) +- **Cowork** - Complete difficult tasks with Claude Cowork +- **Auto mode** - Let the agent decide when to ask before making changes + +## Disconnect Claude + +Open Ollama **Apps** and turn Claude off. Ollama restores Claude's previous configuration. If Claude is open, Ollama asks to restart it so the change takes effect. + +You can also restore Claude from the terminal: ```shell ollama launch claude-desktop --restore ``` -Use [Claude Code](/integrations/claude-code) for Anthropic-compatible coding workflows with Ollama. +Quitting Ollama while Claude is connected also restores Claude's usual configuration. diff --git a/docs/integrations/codex.mdx b/docs/integrations/codex.mdx index b8723f3..93ff3b0 100644 --- a/docs/integrations/codex.mdx +++ b/docs/integrations/codex.mdx @@ -76,3 +76,15 @@ Then run: ``` codex --profile ollama-launch ``` + +## Web search + +Codex web-search requests sent through the Ollama profile are executed by +Ollama for both local and cloud models. Sign in with `ollama signin` to use the +web-search service. + +To disable web search for a Codex session: + +```shell +codex --profile ollama-launch -c 'web_search="disabled"' +``` diff --git a/docs/integrations/deepseek-harness.mdx b/docs/integrations/deepseek-harness.mdx new file mode 100644 index 0000000..c09ff98 --- /dev/null +++ b/docs/integrations/deepseek-harness.mdx @@ -0,0 +1,50 @@ +--- +title: DeepSeek Harness +--- + +[DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) is an open-source coding agent. + +DeepSeek Harness is currently a developer preview. Its upstream configuration may change between releases. + +## Setup + +```shell +ollama launch dsh +``` + +Ollama installs `@deepseek-ai/dsh` if needed. To choose a model: + +```shell +ollama launch dsh --model qwen3.8 +ollama launch dsh --model deepseek-v4-flash:cloud +``` + +To configure without starting: + +```shell +ollama launch dsh --config +``` + +## Web search + +Web search is enabled automatically. It requires Ollama cloud access and a model that supports tools. Run `ollama signin` if needed. + +## Configuration + +Ollama stores its settings in `~/.ollama/launch/dsh/settings.yaml`. These settings load last and set the model, provider, and web search connection. Repeated launches preserve other settings in this file. Ollama does not change `~/.dsh/settings.yaml`, profiles, sessions, or credentials. + +Launch rejects additional `--patch` arguments. Pass other Harness arguments after `--`: + +```shell +ollama launch dsh -- --port 3081 +``` + +## Manual install + +DeepSeek Harness requires Node.js. To install it manually: + +```shell +npm install -g @deepseek-ai/dsh@latest +``` + +Then run `ollama launch dsh`. On Windows, install Node.js for Windows. diff --git a/docs/integrations/index.mdx b/docs/integrations/index.mdx index 1e835e5..c7b67dd 100644 --- a/docs/integrations/index.mdx +++ b/docs/integrations/index.mdx @@ -17,13 +17,21 @@ Run `ollama launch` to see the latest integrations you can run from the terminal Open-source coding agent that edits, runs, and iterates on code. + + + DeepSeek's open-source agent harness with subagents and web search. + ## Connect an assistant -Assistants with memory, skills, and messaging app access. +Use open models in assistant apps. + + Desktop assistant with local and cloud Ollama models. + + Personal assistant for messaging apps and everyday tasks. diff --git a/docs/modelfile.mdx b/docs/modelfile.mdx index 0226d6a..32b6193 100644 --- a/docs/modelfile.mdx +++ b/docs/modelfile.mdx @@ -148,7 +148,7 @@ PARAMETER | -------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | -------------------- | | num_ctx | Sets the size of the context window used to generate the next token. (Default: 2048) | int | num_ctx 4096 | | repeat_last_n | Sets how far back for the model to look back to prevent repetition. (Default: 64, 0 = disabled, -1 = num_ctx) | int | repeat_last_n 64 | -| repeat_penalty | Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. (Default: 1.1) | float | repeat_penalty 1.1 | +| repeat_penalty | Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. (Default: 1.0, disabled) | float | repeat_penalty 1.1 | | temperature | The temperature of the model. Increasing the temperature will make the model answer more creatively. (Default: 0.8) | float | temperature 0.7 | | seed | Sets the random number seed to use for generation. Setting this to a specific number will make the model generate the same text for the same prompt. (Default: 0) | int | seed 42 | | stop | Sets the stop sequences to use. When this pattern is encountered the LLM will stop generating text and return. Multiple stop patterns may be set by specifying multiple separate `stop` parameters in a modelfile. | string | stop "AI assistant:" | diff --git a/docs/openapi.yaml b/docs/openapi.yaml index 34aa9f4..162a906 100644 --- a/docs/openapi.yaml +++ b/docs/openapi.yaml @@ -147,9 +147,12 @@ components: prompt_eval_count: type: integer description: Number of input tokens in the prompt + prompt_eval_cached_count: + type: integer + description: Number of prompt tokens read from the cache prompt_eval_duration: type: integer - description: Time spent evaluating the prompt in nanoseconds + description: Time spent evaluating uncached prompt tokens in nanoseconds eval_count: type: integer description: Number of output tokens generated in the response @@ -191,9 +194,12 @@ components: prompt_eval_count: type: integer description: Number of input tokens in the prompt + prompt_eval_cached_count: + type: integer + description: Number of prompt tokens read from the cache prompt_eval_duration: type: integer - description: Time spent evaluating the prompt in nanoseconds + description: Time spent evaluating uncached prompt tokens in nanoseconds eval_count: type: integer description: Number of output tokens generated in the response @@ -352,9 +358,12 @@ components: prompt_eval_count: type: integer description: Number of tokens in the prompt + prompt_eval_cached_count: + type: integer + description: Number of prompt tokens read from the cache prompt_eval_duration: type: integer - description: Time spent evaluating the prompt in nanoseconds + description: Time spent evaluating uncached prompt tokens in nanoseconds eval_count: type: integer description: Number of tokens generated in the response