Ollama v0.13.3 — Release Notes & Engineering Impact
Ollama v0.13.3. Source highlight: Devstral-Small-2: 24B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.
Ollama archive — Page 3
264 tracked entries. Older history is imported progressively; see source coverage.
Ollama v0.13.2 — Release Notes & Engineering Impact
Ollama v0.13.2. Source highlight: Qwen3-Next: The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.
Ollama v0.13.1 — Release Notes & Engineering Impact
Ollama v0.13.1. Source highlight: Ministral-3: The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware.
Ollama v0.13.0 — Release Notes & Engineering Impact
Ollama v0.13.0. Source highlight: DeepSeek-OCR: DeepSeek-OCR uses optical 2D mapping to compress long contexts, achieving high OCR precision with reduced vision tokens and demonstrating practical value in document processing.
Ollama v0.12.11 — Release Notes & Engineering Impact
Ollama v0.12.11. Source highlight: 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:
Ollama v0.12.10 — Release Notes & Engineering Impact
Ollama v0.12.10. Source highlight: `ollama run` now works with embedding models
Ollama v0.12.9 — Release Notes & Engineering Impact
Ollama v0.12.9. Source highlight: Fix performance regression on CPU-only systems
Ollama v0.12.8 — Release Notes & Engineering Impact
Ollama v0.12.8. Source highlight: `qwen3-vl` performance improvements, including flash attention support by default
Ollama v0.12.7 — Release Notes & Engineering Impact
Ollama v0.12.7. Source highlight: Qwen3-VL: Qwen3-VL is now available in all parameter sizes ranging from 2B to 235B
Ollama v0.12.6 — Release Notes & Engineering Impact
Ollama v0.12.6. Source highlight: Ollama's app now supports searching when running DeepSeek-V3.1, Qwen3 and other models that support tool calling.
Ollama v0.12.5 — Release Notes & Engineering Impact
Ollama v0.12.5. Source highlight: Thinking models now support structured outputs when using the `/api/chat` API
Ollama v0.12.4 — Release Notes & Engineering Impact
Ollama v0.12.4. Source highlight: Flash attention is now enabled by default for Qwen 3 and Qwen 3 Coder
Ollama v0.12.3 — Release Notes & Engineering Impact
Ollama v0.12.3. Source highlight: 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:
Ollama v0.12.2 — Release Notes & Engineering Impact
Ollama v0.12.2. Source highlight: 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 i
Ollama v0.12.1 — Release Notes & Engineering Impact
Ollama v0.12.1. Source highlight: Qwen3 Embedding: state of the art open embedding model by the Qwen team
Ollama v0.12.0 — Release Notes & Engineering Impact
Ollama v0.12.0. Source highlight: Cloud models are now available in preview, allowing you to run a group of larger models with fast, datacenter-grade hardware.
Ollama v0.11.11 — Release Notes & Engineering Impact
Ollama v0.11.11. Source highlight: Improved memory usage when using gpt-oss in Ollama's app
Ollama v0.11.10 — Release Notes & Engineering Impact
Ollama v0.11.10. Source highlight: EmbeddingGemma a new open embedding model that delivers best-in-class performance for its size
Ollama v0.11.9 — Release Notes & Engineering Impact
Ollama v0.11.9. Source highlight: Improved performance via overlapping GPU and CPU computations
Ollama v0.11.8 — Release Notes & Engineering Impact
Ollama v0.11.8. Source highlight: `gpt-oss` now has flash attention enabled by default for systems that support it
Ollama v0.11.7 — Release Notes & Engineering Impact
Ollama v0.11.7. Source highlight: DeepSeek-V3.1 is now available to run via Ollama.
Ollama v0.11.6 — Release Notes & Engineering Impact
Ollama v0.11.6. Source highlight: Ollama's app will now switch between chats faster
Ollama v0.11.5 — Release Notes & Engineering Impact
Ollama v0.11.5. Source highlight: Performance improvements for the `gpt-oss` models
Ollama v0.11.4 — Release Notes & Engineering Impact
Ollama v0.11.4. Source highlight: openai: allow for content _and_ tool calls in the same message
Ollama v0.11.3 — Release Notes & Engineering Impact
Ollama v0.11.3. Source highlight: Fixed issue where `gpt-oss` would consume too much VRAM when split across GPU & CPU or multiple GPUs
Ollama v0.11.2 — Release Notes & Engineering Impact
Ollama v0.11.2. Source highlight: Fix crash in gpt-oss when using kv cache quanitization
Ollama v0.11.0 — Release Notes & Engineering Impact
Ollama v0.11.0. Source highlight: Welcome OpenAI's gpt-oss models
Ollama v0.10.1 — Release Notes & Engineering Impact
Ollama v0.10.1. Source highlight: Fixed unicode character input for Japanese and other languages in Ollama's new app
Ollama v0.10.0 — Release Notes & Engineering Impact
Ollama v0.10.0. Source highlight: Ollama's new app is available for macOS and Windows: Download Ollama
Ollama v0.9.6 — Release Notes & Engineering Impact
Ollama v0.9.6. Source highlight: Fixed styling issue in launch screen
Ollama v0.9.5 — Release Notes & Engineering Impact
Ollama v0.9.5. Source highlight: Updates to Ollama for macOS and Windows
Ollama v0.9.4 — Release Notes & Engineering Impact
Ollama v0.9.4. Source highlight: Updates to Ollama for macOS and Windows
Ollama v0.9.3 — Release Notes & Engineering Impact
Ollama v0.9.3. Source highlight: 
Ollama v0.9.2 — Release Notes & Engineering Impact
Ollama v0.9.2. Source highlight: Fixed issue where tool calls without parameters would not be returned correctly
Ollama v0.9.1 — Release Notes & Engineering Impact
Ollama v0.9.1. Source highlight: Tool calling improvements
Ollama v0.9.0 — Release Notes & Engineering Impact
Ollama v0.9.0. Source highlight: DeepSeek-R1-2508: DeepSeek-R1 has received a minor version upgrade to DeepSeek-R1-0528 for the 8 billion parameter distilled model and the full 671 billion parameter model. In this update, DeepSeek R1 has significantly improved its reasoning and inference capabilities.
Ollama v0.8.0 — Release Notes & Engineering Impact
Ollama v0.8.0. Source highlight: Ollama will now stream responses with tool calls blog post
Ollama v0.7.1 — Release Notes & Engineering Impact
Ollama v0.7.1. Source highlight: Improved model memory management to allocate sufficient memory to prevent crashes when running multimodal models in certain situations
Ollama v0.7.0 — Release Notes & Engineering Impact
Ollama v0.7.0. Source highlight: Ollama now supports multimodal models via Ollama’s new engine, starting with new vision multimodal models:
Ollama v0.6.8 — Release Notes & Engineering Impact
Ollama v0.6.8. Source highlight: Performance improvements for Qwen 3 MoE models (`30b-a3b` and `235b-a22b`) on NVIDIA and AMD GPUs
Ollama v0.6.7 — Release Notes & Engineering Impact
Ollama v0.6.7. Source highlight: Qwen 3: Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.
Ollama v0.6.6 — Release Notes & Engineering Impact
Ollama v0.6.6. Source highlight: IBM Granite 3.3: 2B and 8B models with 128K context length that have been fine-tuned for improved reasoning and instruction-following capabilities.
Ollama v0.6.5 — Release Notes & Engineering Impact
Ollama v0.6.5. Source highlight: Mistral Small 3.1: the best performing vision model in its weight class.
Ollama v0.6.4 — Release Notes & Engineering Impact
Ollama v0.6.4. Source highlight: `/api/show` will now include model capabilities such as `vision`
Ollama v0.6.3 — Release Notes & Engineering Impact
Ollama v0.6.3. Source highlight: New sliding window attention optimizations for Gemma 3, improving inference speed and memory allocation for long context windows.
Ollama v0.6.2 — Release Notes & Engineering Impact
Ollama v0.6.2. Source highlight: Multiple images are now supported in Gemma 3
Ollama v0.6.1 — Release Notes & Engineering Impact
Ollama v0.6.1. Source highlight: Command A: 111 billion parameter model optimized for demanding enterprises that require fast, secure, and high-quality AI.
Ollama v0.6.0 — Release Notes & Engineering Impact
Ollama v0.6.0. Source highlight: Gemma 3: Google Gemma 3 model is now available in 1B, 4B, 12B, and 27B parameter sizes.
Ollama v0.5.13 — Release Notes & Engineering Impact
Ollama v0.5.13. Source highlight: Phi-4-Mini: Phi-4-mini brings significant enhancements in multilingual support, reasoning, and mathematics, and now, the long-awaited function calling feature is finally supported.
Ollama v0.5.12 — Release Notes & Engineering Impact
Ollama v0.5.12. Source highlight: Perplexity R1 1776: A version of the DeepSeek-R1 model that has been post trained to remove its refusal to respond to some sensitive topics.