## Use Ollama models in ChatGPT Desktop Ollama models can now be used directly in ChatGPT Desktop, so you can keep your existing workflow while running open models. Setup is available from the Ollama app on MacOS. This release also improves structured output performance on Apple Silicon, adds support for OpenAI-compatible client tool search and response compaction, and makes images work correctly through compacted responses. **Full Changelog**: https://github.com/ollama/ollama/compare/v0.33.3...v0.34.0

## What's Changed * llm: transcode WebP images for llama-server * renderers/qwen: tolerate non-leading system messages **Full Changelog**: https://github.com/ollama/ollama/compare/v0.32.13...v0.32.14

## What's Changed - Models that don't set a `repeat_penalty` now default to 1.0 (off) instead of 1.1, matching other engines and speeding up speculative decoding; set a per-model parameter if an older model repeats itself. - Faster prefill on NVFP4 MLX models with a global scale, about 7–8% on Qwen3.6 and Muse Glimmer. - Fixed blob verification being skipped when an OCI manifest's config and layer share a digest. ## New Contributors * @vigneshakaviki made their first contribution in https://github.com/ollama/ollama/pull/15504 **Full Changelog**: https://github.com/ollama/ollama/compare/v0.32.8...v0.32.10-rc1

## What's Changed - Qwen3.5 is faster on Apple GPUs: the MLX engine now uses the model's MTP head for speculative decoding automatically - `/v1/chat/completions` streaming now matches OpenAI's wire format: `role` only on the first chunk, `finish_reason` on its own chunk, and usage in a separate chunk with `stream_options.include_usage`. - Truncated OpenAI responses now report `finish_reason: "length"` instead of `"tool_calls"`. - `ollama run kimi-k3` now offers `kimi-k3:cloud` for cloud-only models that publish no default tag, instead of failing. - TUI fixes: pipe-delimited prose no longer renders as a table, Enter accepts the highlighted `@` file completion, and `/prompt` scrolling is no longer laggy. - Experimental image generation has been temporarily removed. Continue using 0.32.5 for image generation support - Updated the MLX and llama.cpp engines. **Full Changelog**: https://github.com/ollama/ollama/compare/v0.32.5...v0.32.6-rc0

## What's Changed - Fixed model downloads that stall before sending data. - Improved integrations: restored Claude Code Channels, fixed Anthropic thinking streams, and made Hermes Desktop respect `--force-build`. - Expanded GPU support with CUDA on Windows ARM64, B200 support through CUDA 12, and lower memory use on Linux CUDA/ROCm iGPUs. - Added chat, thinking, and tool calling support for Laguna 2.1 models, including a Metal inference fix. - Fixed GLM tool calls being silently dropped at the end of generation. - Updated the MLX and llama.cpp engines. **Full Changelog**: https://github.com/ollama/ollama/compare/v0.32.1...v0.32.3

## What's Changed * mlx update by @dhiltgen in https://github.com/ollama/ollama/pull/17332 * model/parsers: finalize incomplete GLM tool calls by @dhiltgen in https://github.com/ollama/ollama/pull/17250 * docs: update retirements by @mxyng in https://github.com/ollama/ollama/pull/17289 * model: align Laguna with upstream llama.cpp by @dhiltgen in https://github.com/ollama/ollama/pull/17335 **Full Changelog**: https://github.com/ollama/ollama/compare/v0.32.2...v0.32.3-rc0

## What's Changed * ci: avoid unbounded parallelism by @dhiltgen in https://github.com/ollama/ollama/pull/16966 * fix cuda toolkit lookup and parallel by @MarkWard0110 in https://github.com/ollama/ollama/pull/16613 * docs(cloud): update retirement list by @mxyng in https://github.com/ollama/ollama/pull/17000 * discover: fall back to standard CUDA when the JetPack runner is absent by @kevinpark1217 in https://github.com/ollama/ollama/pull/16949 * agent: harness core by @ParthSareen in https://github.com/ollama/ollama/pull/16963 * llama: clean up dead code from llama-server work by @dhiltgen in https://github.com/ollama/ollama/pull/17007 * rocm: remove no longer supported devices by @dhiltgen in https://github.com/ollama/ollama/pull/17010 * compat: use UTF-8-safe file open by @dhiltgen in https://github.com/ollama/ollama/pull/16999 * llama: enable FA on CUDA CC 6.x GPUs by @dhiltgen in https://github.com/ollama/ollama/pull/16994 * mlx: x/create rewrite by @pdevine in https://github.com/ollama/ollama/pull/16919 * llama.cpp update by @dhiltgen in https://github.com/ollama/ollama/pull/17055 * server: remove OLLAMA_EXPERIMENT=client2 by @mxyng in https://github.com/ollama/ollama/pull/16962 * mlx: update to de7b4ed9 by @dhiltgen in https://github.com/ollama/ollama/pull/17056 ## New Contributors * @kevinpark1217 made their first contribution in https://github.com/ollama/ollama/pull/16949 **Full Changelog**: https://github.com/ollama/ollama/compare/v0.31.1...v0.31.2-rc0

## What's Changed * mlx: tighten up gemma4 moe loading code by @pdevine in https://github.com/ollama/ollama/pull/16964 * mlx: bump to latest version to include new small batch matmul kernel @jessegross @dhiltgen * llama.cpp: bump to b9840 @dhiltgen * improved gemma4 MTP performance @jessegross **Full Changelog**: https://github.com/ollama/ollama/compare/v0.31.0...v0.31.1

## What's Changed * docs: update Zod example to use native toJSONSchema by @fuleinist in https://github.com/ollama/ollama/pull/14746 * launch: use native Windows Hermes config path by @BruceMacD in https://github.com/ollama/ollama/pull/16558 ## New Contributors * @fuleinist made their first contribution in https://github.com/ollama/ollama/pull/14746 **Full Changelog**: https://github.com/ollama/ollama/compare/v0.30.6...v0.30.7-rc1

## What's Changed * mlx: refined model push behavior by @dhiltgen in https://github.com/ollama/ollama/pull/15431 * test: integration test hardening by @dhiltgen in https://github.com/ollama/ollama/pull/13532 * app: harden update flows by @dhiltgen in https://github.com/ollama/ollama/pull/16100 **Full Changelog**: https://github.com/ollama/ollama/compare/v0.23.2...v0.23.3-rc0