Ollama v0.5.11 — Release Notes & Engineering Impact
Ollama v0.5.11. Source highlight: Fixed `The system cannot find the path specified` errors when running models in some cases on Windows
Ollama archive — Page 4
264 tracked entries. Older history is imported progressively; see source coverage.
Ollama v0.5.10 — Release Notes & Engineering Impact
Ollama v0.5.10. Source highlight: Fixed issue on multi-GPU Windows and Linux machines where memory estimations would be incorrect
Ollama v0.5.9 — Release Notes & Engineering Impact
Ollama v0.5.9. Source highlight: DeepScaleR: A fine-tuned version of Deepseek-R1-Distilled-Qwen-1.5B that surpasses the performance of OpenAI’s o1-preview with just 1.5B parameters on popular math evaluations.
Ollama v0.5.8 — Release Notes & Engineering Impact
Ollama v0.5.8. Source highlight: Ollama will now use AVX-512 instructions where available for additional CPU acceleration
Ollama v0.5.7 — Release Notes & Engineering Impact
Ollama v0.5.7. Source highlight: Fixed issue where using two `FROM` commands in `Modelfile`
Ollama v0.5.6 — Release Notes & Engineering Impact
Ollama v0.5.6. Source highlight: Fixed errors that would occur when running `ollama create` on Windows and when using absolute paths
Ollama v0.5.5 — Release Notes & Engineering Impact
Ollama v0.5.5. Source highlight: Phi-4: Phi 4 is a 14B parameter, state-of-the-art open model from Microsoft.
Ollama v0.5.4 — Release Notes & Engineering Impact
Ollama v0.5.4. Source highlight: Falcon3: A family of efficient AI models under 10B parameters performant in science, math, and coding through innovative training techniques.
Ollama v0.5.3 — Release Notes & Engineering Impact
Ollama v0.5.3. Source highlight: Fixed runtime errors on older Intel Macs
Ollama v0.5.2 — Release Notes & Engineering Impact
Ollama v0.5.2. Source highlight: EXAONE 3.5: a collection of instruction-tuned bilingual (English and Korean) generative models ranging from 2.4B to 32B parameters, developed and released by LG AI Research.
Ollama v0.5.1 — Release Notes & Engineering Impact
Ollama v0.5.1. Source highlight: Fixed issue where Ollama's API would generate JSON output when specifying `"format": null`
Ollama v0.5.0 — Release Notes & Engineering Impact
Ollama v0.5.0. Source highlight: Llama 3.3: a new state of the art 70B model. Llama 3.3 70B offers similar performance compared to Llama 3.1 405B model.
Ollama v0.4.7 — Release Notes & Engineering Impact
Ollama v0.4.7. Source highlight: The OpenAI compatible chat endpoint now includes index fields for tool calls, fixing issue when using `chat.completions.stream` and other methods
Ollama v0.4.6 — Release Notes & Engineering Impact
Ollama v0.4.6. Source highlight: QwQ: an experimental research model by the Qwen team, focused on advancing AI reasoning capabilities.
Ollama v0.4.5 — Release Notes & Engineering Impact
Ollama v0.4.5. Source highlight: Ollama Python library improvements
Ollama v0.4.4 — Release Notes & Engineering Impact
Ollama v0.4.4. Source highlight: Marco-o1: An open large reasoning model for real-world solutions by the Alibaba International Digital Commerce Group (AIDC-AI).
Ollama v0.4.3 — Release Notes & Engineering Impact
Ollama v0.4.3. Source highlight: Tülu 3: Tülu 3 is a leading instruction following model family, offering fully open-source data, code, and recipes by the The Allen Institute for AI.
Ollama v0.4.2 — Release Notes & Engineering Impact
Ollama v0.4.2. Source highlight: Ollama now supports NVIDIA Jetson. To install Ollama on NVIDIA Jetson devices, run:
Ollama v0.4.1 — Release Notes & Engineering Impact
Ollama v0.4.1. Source highlight: Fixed an issue where the cuda v12 runner on linux was incorrectly linked
Ollama v0.4.0 — Release Notes & Engineering Impact
Ollama v0.4.0. Source highlight: Ollama 0.4 adds support for Llama 3.2 Vision. After upgrading or downloading Ollama, run:
Ollama v0.3.14 — Release Notes & Engineering Impact
Ollama v0.3.14. Source highlight: Granite 3 MoE: The IBM Granite 1B and 3B models are the first mixture of experts (MoE) Granite models from IBM designed for low latency usage.
Ollama v0.3.13 — Release Notes & Engineering Impact
Ollama v0.3.13. Source highlight: Llama Guard 3: a series of models by Meta, fine-tuned for content safety classification of LLM inputs and responses.
Ollama v0.3.12 — Release Notes & Engineering Impact
Ollama v0.3.12. Source highlight: Llama 3.2: Meta's Llama 3.2 goes small with 1B and 3B models.
Ollama v0.3.11 — Release Notes & Engineering Impact
Ollama v0.3.11. Source highlight: Solar-Pro-Preview: an advanced large language model (LLM) with 22 billion parameters designed to fit into a single GPU.
Ollama v0.3.10 — Release Notes & Engineering Impact
Ollama v0.3.10. Source highlight: MiniCPM-V: A powerful, multi-modal model with leading performance on several benchmarks.
Ollama v0.3.9 — Release Notes & Engineering Impact
Ollama v0.3.9. Source highlight: Fixed error that would occur when running Ollama on Linux machines with the ARM architecture
Ollama v0.3.8 — Release Notes & Engineering Impact
Ollama v0.3.8. Source highlight: Fixed error where the `ollama` CLI couldn't be found on the path when upgrading Ollama on Windows
Ollama v0.3.7 — Release Notes & Engineering Impact
Ollama v0.3.7. Source highlight: Hermes 3: Hermes 3 is the latest version of the flagship Hermes series of LLMs by Nous Research, which includes support for tool calling.
Ollama v0.3.6 — Release Notes & Engineering Impact
Ollama v0.3.6. Source highlight: Fixed issue where `/api/embed` would return an error instead of loading the model when the `input` field was not provided.
Ollama v0.3.5 — Release Notes & Engineering Impact
Ollama v0.3.5. Source highlight: Fixed `Incorrect function` error when downloading models on Windows
Ollama v0.3.4 — Release Notes & Engineering Impact
Ollama v0.3.4. Source highlight: New embedding models
Ollama v0.3.3 — Release Notes & Engineering Impact
Ollama v0.3.3. Source highlight: The `/api/embed` endpoint now returns statistics: `total_duration`, `load_duration`, and `prompt_eval_count`
Ollama v0.3.2 — Release Notes & Engineering Impact
Ollama v0.3.2. Source highlight: Fixed issue where `ollama pull` would not resume download progress
Ollama v0.3.1 — Release Notes & Engineering Impact
Ollama v0.3.1. Source highlight: Google's Gemma 2 has a new 2B parameter small model!
Ollama v0.3.0 — Release Notes & Engineering Impact
Ollama v0.3.0. Source highlight: Ollama now supports tool calling with popular models such as Llama 3.1. This enables a model to answer a given prompt using tool(s) it knows about, making it possible for models to perform more complex tasks or interact with the outside world.
Ollama v0.2.8 — Release Notes & Engineering Impact
Ollama v0.2.8. Source highlight: Mistral Nemo: A state-of-the-art 12B model with 128k context length, built by Mistral AI in collaboration with NVIDIA.
Ollama v0.2.7 — Release Notes & Engineering Impact
Ollama v0.2.7. Source highlight: Fixed issue where last message when streaming would omit the `content` response field
Ollama v0.2.6 — Release Notes & Engineering Impact
Ollama v0.2.6. Source highlight: Mathstral: MathΣtral is a 7B model designed for math reasoning and scientific discovery by Mistral AI.
Ollama v0.2.5 — Release Notes & Engineering Impact
Ollama v0.2.5. Source highlight: Fixed issue where a model's `SYSTEM` message not be applied
Ollama v0.2.4 — Release Notes & Engineering Impact
Ollama v0.2.4. Source highlight: Fixed issue where `context`, `load_duration` and `total_duration` fields would not be set in the `/api/generate` endpoint.
Ollama v0.2.3 — Release Notes & Engineering Impact
Ollama v0.2.3. Source highlight: Fix issue where system prompt would not be applied
Ollama v0.2.2 — Release Notes & Engineering Impact
Ollama v0.2.2. Source highlight: Fixed errors that occurred when using Ollama with Nvidia V100 GPUs
Ollama v0.2.1 — Release Notes & Engineering Impact
Ollama v0.2.1. Source highlight: Fixed issue where setting `OLLAMA_NUM_PARALLEL` would cause models to be reloaded after each request
Ollama v0.2.0 — Release Notes & Engineering Impact
Ollama v0.2.0. Source highlight: Ollama 0.2.0 is now available with concurrency support. This unlocks 2 specific features:
Ollama v0.1.48 — Release Notes & Engineering Impact
Ollama v0.1.48. Source highlight: Fixed issue where Gemma 2 would continuously output when reaching context limits
Ollama v0.1.47 — Release Notes & Engineering Impact
Ollama v0.1.47. Source highlight: Added support for Google Gemma 2 models (9B and 27B)
Ollama v0.1.46 — Release Notes & Engineering Impact
Ollama v0.1.46. Source highlight: Increased model loading speed with `ollama run`, especially if running an already-loaded model
Ollama v0.1.45 — Release Notes & Engineering Impact
Ollama v0.1.45. Source highlight: DeepSeek-Coder-V2: A 16B & 236B open-source Mixture-of-Experts code language model that achieves performance comparable to GPT4-Turbo in code-specific tasks.
Ollama v0.1.44 — Release Notes & Engineering Impact
Ollama v0.1.44. Source highlight: Fixed issue where unicode characters such as emojis would not be loaded correctly when running `ollama create`
Ollama v0.1.43 — Release Notes & Engineering Impact
Ollama v0.1.43. Source highlight: New import.md guide for converting and importing models to Ollama