Open-source framework that allows you to download, set up, and run Large Language Models (LLMs)—like Llama, Mistral, and DeepSeek
Update Feb 19, 2026 tracked by Updatify
ollama launch cline added for the Cline CLI ollama launch <integration> will now always show the model picker Full Changelog: https://github.com/ollama/ollama/compare/v0.16.2…v0.16.3
Update Feb 14, 2026 tracked by Updatify
ollama launch claude now supports searching the web when using :cloud models ollama in PowerShell ollama serve manually, set OLLAMA_NO_CLOUD=1. Full Changelog: https://github.com/ollama/ollama/compare/v0.16.1…v0.16.2-rc0
Update Feb 12, 2026 tracked by Updatify
curl install script on macOS will now only prompt for your password if its required iem install script in Windows will now show progress OLLAMA_LOAD_TIMEOUT variable Full Changelog: https://github.com/ollama/ollama/compare/v0.16.0…v0.16.1
Update Feb 12, 2026 tracked by Updatify
ollama
The new ollama command makes it easy to launch your favorite apps with models using Ollama
ollama launch pi Full Changelog: https://github.com/ollama/ollama/compare/v0.15.6…v0.16.0
Update Feb 7, 2026 tracked by Updatify
ollama launch droid ollama launch will now download missing models instead of erroring ollama launch claude would cause context compaction when providing images Full Changelog: https://github.com/ollama/ollama/compare/v0.15.5…v0.15.6
Update Feb 3, 2026 tracked by Updatify
ollama launchollama launch can now be provided arguments, for example ollama launch claude -- --resume ollama launch will now work run subagents when using ollama launch claude ollama launch opencode ollama launch for planning, deep research, and similar tasks ollama signin will now open a browser window to make signing in easier ollama signin will now open the browser to the connect page num_predict in the API num_predict Full Changelog: https://github.com/ollama/ollama/compare/v0.15.4…v0.15.5
Update Feb 1, 2026 tracked by Updatify
ollama launch openclaw will now enter the standard OpenClaw onboarding flow if this has not yet been completed. Full Changelog: https://github.com/ollama/ollama/compare/v0.15.3…v0.15.4
Update Feb 1, 2026 tracked by Updatify
ollama launch clawdbot to ollama launch openclaw to reflect the project’s new name ollama launch will now use the value of OLLAMA_HOST when running it Full Changelog: https://github.com/ollama/ollama/compare/v0.15.2…v0.15.3
Update Jan 27, 2026 tracked by Updatify
ollama launch clawdbot command for launching Clawdbot using Ollama models Full Changelog: https://github.com/ollama/ollama/compare/v0.15.1…v0.15.2
Update Jan 24, 2026 tracked by Updatify
ollama launch would not detect claude and would incorrectly update opencode configurations Full Changelog: https://github.com/ollama/ollama/compare/v0.15.0…v0.15.1
Update Jan 21, 2026 tracked by Updatify
ollama launch
A new ollama launch command to use Ollama’s models with Claude Code, Codex, OpenCode, and Droid without separate configuration.
ollama launch command for Claude Code, Codex, OpenCode, and Droid """ would not work when using ollama run ollama run Update Jan 16, 2026 tracked by Updatify
ollama create and ollama show commands for experimental models /api/generate API can now be used for image generation ollama rm would only stop the first model in the list if it were running Full Changelog: https://github.com/ollama/ollama/compare/v0.14.2…v0.14.3
Update Jan 16, 2026 tracked by Updatify
/v1/responses API to better confirm to OpenResponses specification Full Changelog: https://github.com/ollama/ollama/compare/v0.14.1…v0.14.2
Update Jan 14, 2026 tracked by Updatify
Experimental image generation models are available for macOS and Linux (CUDA) in Ollama:
ollama run x/z-image-turbo
Note:
xis a username on ollama.com where experimental models are uploaded
More models coming soon:
Full Changelog: https://github.com/ollama/ollama/compare/v0.14.0…v0.14.1
Update Jan 10, 2026 tracked by Updatify
ollama run --experimental CLI will now open a new Ollama CLI that includes an agent loop and the bash tool /v1/messages API REQUIRES command for the Modelfile allows declaring which version of Ollama is required for the model NaN or -Inf zst compression Full Changelog: https://github.com/ollama/ollama/compare/v0.13.5…v0.14.0-rc2
Update Dec 18, 2025 tracked by Updatify
bert architecture models now run on Ollama’s engine Full Changelog: https://github.com/ollama/ollama/compare/v0.13.4…v0.13.5
Update Dec 13, 2025 tracked by Updatify
Full Changelog: https://github.com/ollama/ollama/compare/v0.13.3…v0.13.4-rc0
Update Dec 9, 2025 tracked by Updatify
/api/embed and /v1/embeddings Full Changelog: https://github.com/ollama/ollama/compare/v0.13.2…v0.13.3
Update Dec 4, 2025 tracked by Updatify
mistral-3, gemma3, qwen3-vl and more. This improves memory utilization and performance when providing images as input. deepseek-v3.1 would always think even with thinking is disabled in Ollama’s app Full Changelog: https://github.com/ollama/ollama/compare/v0.13.1…v0.13.2
Update Nov 27, 2025 tracked by Updatify
nomic-embed-text will now use Ollama’s engine by default cogito-v2.1 Unmarshal: errors Full Changelog: https://github.com/ollama/ollama/compare/v0.13.0…v0.13.1
Update Nov 19, 2025 tracked by Updatify
DeepSeek-OCR is now available on Ollama. Example inputs:
ollama run deepseek-ocr "/path/to/image\n<|grounding|>Given the layout of the image."
ollama run deepseek-ocr "/path/to/image\nFree OCR."
ollama run deepseek-ocr "/path/to/image\nParse the figure."
ollama run deepseek-ocr "/path/to/image\nExtract the text in the image."
ollama run deepseek-ocr "/path/to/image\n<|grounding|>Convert the document to markdown."
bench tool
Ollama’s GitHub repo now includes a bench tool that can be used to test model performance. For the time being this is a separate tool that can be built in the Ollama GitHub repository:
First, install Go. Then from the root of the Ollama repository run:
go run ./cmd/bench -model gpt-oss:20b
For more information see the tool’s documentation
Full Changelog: https://github.com/ollama/ollama/compare/v0.12.11…v0.13.0
Update Nov 12, 2025 tracked by Updatify
Ollama’s API and OpenAI-compatible API now support log probabilities. Log probabilities of output tokens indicate the likelihood of each token occurring in the sequence given the context. This is useful for different use cases:
To enable Logprobs, provide "logprobs": true to Ollama’s API:
curl http://localhost:11434/api/generate -d '{
"model": "gemma3",
"prompt": "Why is the sky blue?",
"logprobs": true
}'
When log probabilities are requested, response chunks will now include a "logprobs" field with the token, log probability and raw bytes (for partial unicode).
{
"model": "gemma3",
"created_at": "2025-11-14T22:17:56.598562Z",
"response": "Okay",
"done": false,
"logprobs": [
{
"token": "Okay",
"logprob": -1.3434503078460693,
"bytes": [
79,
107,
97,
121
]
}
]
}
top_logprobs
When setting "top_logprobs", a number of most-likely tokens are also provided, making it possible to introspect alternative tokens. Below is an example request.
curl http://localhost:11434/api/generate -d '{
"model": "gemma3",
"prompt": "Why is the sky blue?",
"logprobs": true,
"top_logprobs": 3
}'
This will generate a stream of response chunks with the following fields:
{
"model": "gemma3",
"created_at": "2025-11-14T22:26:10.466324Z",
"response": "The",
"done": false,
"logprobs": [
{
"token": "The",
"logprob": -0.8361086845397949,
"bytes": [
84,
104,
101
],
"top_logprobs": [
{
"token": "The",
"logprob": -0.8361086845397949,
"bytes": [
84,
104,
101
]
},
{
"token": "Okay",
"logprob": -1.2590975761413574,
"bytes": [
79,
107,
97,
121
]
},
{
"token": "That",
"logprob": -1.2686877250671387,
"bytes": [
84,
104,
97,
116
]
}
]
}
]
}
Thank you @baptistejamin for adding Logprobs to Ollama’s API.
Ollama 0.12.11 includes support for Vulkan acceleration. Vulkan brings support for a broad range of GPUs from AMD, Intel, and iGPUs. Vulkan support is not yet enabled by default, and requires opting in by running Ollama with a custom environment variable:
OLLAMA_VULKAN=1 ollama serve
On Powershell, use:
$env:OLLAMA_VULKAN="1"
ollama serve
For issues or feedback on using Vulkan with Ollama, create an issue labelled Vulkan and make sure to include server logs where possible to aid in debugging.
"required" field in tool definitions will now be omitted if not specified "tool_call_id" would be omitted when using the OpenAI-compatible API. ollama create would import data from both consolidated.safetensors and other safetensor files. OLLAMA_VULKAN=1. For example: OLLAMA_VULKAN=1 ollama serve Full Changelog: https://github.com/ollama/ollama/compare/v0.12.10…v0.12.11
Update Nov 5, 2025 tracked by Updatify
ollama run now works with embedding models
ollama run can now run embedding models to generate vector embeddings from text:
ollama run embeddinggemma "Hello world"
Content can also be provided to ollama run via standard input:
echo "Hello world" | ollama run embeddinggemma
qwen3-vl:235b and qwen3-vl:235b-instruct /api/chat API ollama run now works with embedding models Full Changelog: https://github.com/ollama/ollama/compare/v0.12.9…v0.12.10
Update Oct 31, 2025 tracked by Updatify
Full Changelog: https://github.com/ollama/ollama/compare/v0.12.8…v0.12.9
Update Oct 30, 2025 tracked by Updatify
qwen3-vl performance improvements, including flash attention support by default qwen3-vl will now output less leading whitespace in the response when thinking deepseek-v3.1 thinking could not be disabled in Ollama’s new app qwen3-vl would fail to interpret images with transparent backgrounds ollama rm Full Changelog: https://github.com/ollama/ollama/compare/v0.12.7…v0.12.8
Update Oct 29, 2025 tracked by Updatify
Ollama’s new app now includes a way to add one or many files when prompting the model:
For better responses, thinking levels can now be adjusted for the gpt-oss models:
New API documentation is available for Ollama’s API: https://docs.ollama.com/api
embeddinggemma /v1/embeddings endpoint now supports encoding_format parameter {"name": name, "arguments": args} (thanks @rick-github!) FROM <model> would not inherit RENDERER or PARSER commands Full Changelog: https://github.com/ollama/ollama/compare/v0.12.6…v0.12.7
Update Oct 15, 2025 tracked by Updatify
qwen3-coder would act in raw mode when using /api/generate or ollama run qwen3-coder <prompt> qwen3-embedding providing invalid results num_gpu is set tool_index with a value of 0 would not be sent to the model Experimental support for Vulkan is now available when you build locally from source. This will enable additional GPUs from AMD, and Intel which are not currently supported by Ollama. To build locally, install the Vulkan SDK and set VULKAN_SDK in your environment, then follow the developer instructions. In a future release, Vulkan support will be included in the binary release as well. Please file issues if you run into any problems.
Full Changelog: https://github.com/ollama/ollama/compare/v0.12.5…v0.12.6
Update Oct 10, 2025 tracked by Updatify
/api/chat API "think": false would show an error instead of being silently ignored deepseek-r1 output issues Full Changelog: https://github.com/ollama/ollama/compare/v0.12.4…v0.12.5-rc0
Update Oct 3, 2025 tracked by Updatify
keep_alive in the API would accept different values for the /api/chat and /api/generate endpoints qwen3-coder OLLAMA_FLASH_ATTENTION can now be overridden to 0 for models that have flash attention enabled by default Full Changelog: https://github.com/ollama/ollama/compare/v0.12.3…v0.12.4-rc3
Update Sep 26, 2025 tracked by Updatify
DeepSeek-V3.1-Terminus: DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode. It delivers more stable & reliable outputs across benchmarks compared to the previous version:
Run on Ollama’s cloud:
ollama run deepseek-v3.1:671b-cloudRun locally (requires 500GB+ of VRAM)
ollama run deepseek-v3.1Kimi-K2-Instruct-0905: Kimi K2-Instruct-0905 is the latest, most capable version of Kimi K2. It is a state-of-the-art mixture-of-experts (MoE) language model, featuring 32 billion activated parameters and a total of 1 trillion parameters.
ollama run kimi-k2:1t-cloudollama push will now provide a URL to follow to sign in /load would crash Full Changelog: https://github.com/ollama/ollama/compare/v0.12.2…v0.12.3
Update Sep 24, 2025 tracked by Updatify
A new web search API is now available in Ollama. Ollama provides a generous free tier of web searches for individuals to use, and higher rate limits are available via Ollama’s cloud. This web search capability can augment models with the latest information from the web to reduce hallucinations and improve accuracy.
Full Changelog: https://github.com/ollama/ollama/compare/v0.12.1…v0.12.2
Update Sep 21, 2025 tracked by Updatify
& characters in Qwen3-Coder would not be parsed correctly when function calling ollama signin would not work properly on Linux Full Changelog: https://github.com/ollama/ollama/compare/v0.12.0…v0.12.1
Update Sep 18, 2025 tracked by Updatify
Cloud models are now available in preview, allowing you to run a group of larger models with fast, datacenter-grade hardware.
To run a cloud model, use:
ollama run qwen3-coder:480b-cloud
ollama create <think> if provided in the prompt for /api/generate by @rick-github Full Changelog: https://github.com/ollama/ollama/compare/v0.11.11…v0.12.0
Update Sep 11, 2025 tracked by Updatify
dimensions field to embed requests by @mxyng in https://github.com/ollama/ollama/pull/12242 Full Changelog: https://github.com/ollama/ollama/compare/v0.11.10…v0.11.11
Update Sep 4, 2025 tracked by Updatify
Full Changelog: https://github.com/ollama/ollama/compare/v0.11.9…v0.11.10
Update Sep 2, 2025 tracked by Updatify
Full Changelog: https://github.com/ollama/ollama/compare/v0.11.8…v0.11.9-rc0
Update Aug 27, 2025 tracked by Updatify
gpt-oss now has flash attention enabled by default for systems that support it gpt-oss Full Changelog: https://github.com/ollama/ollama/compare/v0.11.7…v0.11.8
Update Aug 25, 2025 tracked by Updatify
DeepSeek-V3.1 is now available to run via Ollama.
This model supports hybrid thinking, meaning thinking can be enabled or disabled by setting think in Ollama’s API:
curl http://localhost:11434/api/chat -d '{
"model": "deepseek-v3.1",
"messages": [
{
"role": "user",
"content": "why is the sky blue?"
}
],
"think": true
}'
In Ollama’s CLI, thinking can be enabled or disabled by running the /set think or /set nothink commands.
DeepSeek-V3.1 has over 671B parameters, and so a large amount of VRAM is required to run it. Ollama’s Turbo mode (in preview) provides access to powerful hardware in the cloud you can use to run the model.
deepseek-v3.1:671b from the model selector OLLAMA_HOST=ollama.com ollama run deepseek-v3.1
For instructions on using Turbo with Ollama’s Python and JavaScript library, see the docs
<think> tag (e.g. DeepSeek-V3.1) <think> tag from a model { or } would not be parsed correctly Full Changelog: https://github.com/ollama/ollama/compare/v0.11.6…v0.11.7
Update Aug 20, 2025 tracked by Updatify
Full Changelog: https://github.com/ollama/ollama/compare/v0.11.5…v0.11.6
Update Aug 15, 2025 tracked by Updatify
gpt-oss models OLLAMA_NEW_ESTIMATES=1 ollama serve and will soon be enabled by default. OLLAMA_FLASH_ATTENTION=1 will also enable flash attention for pure-CPU models reasoning_effort Full Changelog: https://github.com/ollama/ollama/compare/v0.11.4…v0.11.5
Update Aug 7, 2025 tracked by Updatify
Full Changelog: https://github.com/ollama/ollama/compare/v0.11.3…v0.11.4
Update Aug 6, 2025 tracked by Updatify
gpt-oss would consume too much VRAM when split across GPU & CPU or multiple GPUs Full Changelog: https://github.com/ollama/ollama/compare/v0.11.2…v0.11.3
Update Aug 5, 2025 tracked by Updatify
Full Changelog: https://github.com/ollama/ollama/compare/v0.11.1…v0.11.2
Update Aug 5, 2025 tracked by Updatify
Ollama partners with OpenAI to bring its latest state-of-the-art open weight models to Ollama. The two models, 20B and 120B, bring a whole new local chat experience, and are designed for powerful reasoning, agentic tasks, and versatile developer use cases.
OpenAI utilizes quantization to reduce the memory footprint of the gpt-oss models. The models are post-trained with quantization of the mixture-of-experts (MoE) weights to MXFP4 format, where the weights are quantized to 4.25 bits per parameter. The MoE weights are responsible for 90+% of the total parameter count, and quantizing these to MXFP4 enables the smaller model to run on systems with as little as 16GB memory, and the larger model to fit on a single 80GB GPU.
Ollama is supporting the MXFP4 format natively without additional quantizations or conversions. New kernels are developed for Ollama’s new engine to support the MXFP4 format.
Ollama collaborated with OpenAI to benchmark against their reference implementations to ensure Ollama’s implementations have the same quality.
You can get started by downloading the latest Ollama version (v0.11)
The model can be downloaded directly in Ollama’s new app or via the terminal:
ollama run gpt-oss:20b
ollama run gpt-oss:120b
Full Changelog: https://github.com/ollama/ollama/compare/v0.10.1…v0.11.0
Update Jul 31, 2025 tracked by Updatify
ollama serve Full Changelog: https://github.com/ollama/ollama/compare/v0.10.0…v0.10.1
Update Jul 18, 2025 tracked by Updatify
Ollama’s new app is available for macOS and Windows: Download Ollama
ollama ps will now show the context length of loaded models gemma3n models by 2-3x granite3.3 and mistral-nemo models add and get_address ollama show would report an error ollama run will more gracefully display errors Full Changelog: https://github.com/ollama/ollama/compare/v0.9.6…v0.10.0
Update Jul 8, 2025 tracked by Updatify
tool_name can now be provided in messages with "role": "tool" using the /api/chat endpoint Full Changelog: https://github.com/ollama/ollama/compare/v0.9.5…v0.9.6-rc0
Update Jul 2, 2025 tracked by Updatify
A new version of Ollama’s macOS and Windows applications are now available. New improvements to the apps will be introduced over the coming releases:
Ollama can now be exposed on the network, allowing others to access Ollama on other devices or even over the internet. This is useful for having Ollama running on a powerful Mac, PC or Linux computer while making it accessible to less powerful devices.
The directory in which models are stored can now be modified! This allows models to be stored on external hard disks or alternative directories than the default.
The macOS app is now a native application and starts much faster while requiring a much smaller installation.
ollama CLI would not be installed by Ollama on macOS on startup ollama-darwin.tgz were not notarized Update Jun 27, 2025 tracked by Updatify
A new version of Ollama’s macOS and Windows applications are now available. New improvements to the apps will be introduced over the coming releases:
Ollama can now be exposed on the network, allowing others to access Ollama on other devices or even over the internet. This is useful for having Ollama running on a powerful Mac, PC or Linux computer while making it accessible to less powerful devices.
The directory in which models are stored can now be modified! This allows models to be stored on external hard disks or alternative directories than the default.
The macOS app is now a native application and starts much faster while requiring a much smaller installation.
Full Changelog: https://github.com/ollama/ollama/compare/v0.9.3…v0.9.4
Update Jun 25, 2025 tracked by Updatify
Ollama now supports Gemma 3n.
Gemma 3n models are designed for efficient execution on everyday devices such as laptops, tablets or phones. These models were trained with data in over 140 spoken languages.
ollama run gemma3n:e2b
ollama run gemma3n:e4b
Full Changelog: https://github.com/ollama/ollama/compare/v0.9.2…v0.9.3