{"version":"https://jsonfeed.org/version/1.1","title":"Ollama — Lab Ledger Desk","home_page_url":"https://labledgerdesk.pages.dev/lab/ollama/","feed_url":"https://labledgerdesk.pages.dev/lab/ollama/feed.json","description":"Open-source Ollama releases filed by Lab Ledger Desk.","items":[{"id":"https://labledgerdesk.pages.dev/b/2026/11/16/v0-40-17-rc1-mlxrunner-batch-speculative-verification-19663/","url":"https://labledgerdesk.pages.dev/b/2026/11/16/v0-40-17-rc1-mlxrunner-batch-speculative-verification-19663/","title":"Ollama v0.40.17","content_text":"What's Changed Improve ChatGPT CLI restore and graceful restart by @hoyyeva in #19583 cmd/launch: use thinking metadata for OpenCode settings by @hoyyeva in #19409 llm: use mmap for Vulkan integrated GPUs by @dhiltgen in #19640 rpreslar4765/Ollama by @rpreslar4765 in #17163 mlx: support concurrent generation by…","date_published":"2026-11-16T16:43:54.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.17-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/10/ollama-v0-40-17/","url":"https://labledgerdesk.pages.dev/b/2026/11/10/ollama-v0-40-17/","title":"Ollama v0.40.17","content_text":"What's Changed Improve ChatGPT CLI restore and graceful restart by @hoyyeva in #19583 cmd/launch: use thinking metadata for OpenCode settings by @hoyyeva in #19409 llm: use mmap for Vulkan integrated GPUs by @dhiltgen in #19640 rpreslar4765/Ollama by @rpreslar4765 in #17163 mlx: support concurrent generation by…","date_published":"2026-11-10T22:25:00.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.17-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/08/v0-40-16-rc0-add-ripple-preview-support-to-the-mlx-runner-19592/","url":"https://labledgerdesk.pages.dev/b/2026/11/08/v0-40-16-rc0-add-ripple-preview-support-to-the-mlx-runner-19592/","title":"v0.40.16-rc0: Add ripple preview support to the MLX runner (#19592)","content_text":"Add the Ripple model, renderer, parser and import registration. Support per-layer NoPE in DFlash drafts. Fuse MLP activation, value blending and attention gating. Add reference fixtures and model/runner regression tests.","date_published":"2026-11-08T00:36:53.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.16-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/08/v0-40-16-add-ripple-preview-support-to-the-mlx-runner-19592/","url":"https://labledgerdesk.pages.dev/b/2026/11/08/v0-40-16-add-ripple-preview-support-to-the-mlx-runner-19592/","title":"Ollama v0.40.16","content_text":"Ripple Preview Ripple Preview from Ricursive Intelligence is a 65B dense model built for agentic coding. …","date_published":"2026-11-08T00:36:53.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.16"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/05/ollama-v0-40-15/","url":"https://labledgerdesk.pages.dev/b/2026/11/05/ollama-v0-40-15/","title":"Ollama v0.40.15","content_text":"MiMo V2.6 Xiaomi MiMo's mimo-v2.6 , a 9B model built for code and agent workflows, is now supported in Ollama. ollama run mimo-v2.6 AutoTrust decision models AutoTrust decision models read text and return choices, yes or no probabilities, and ratings for the questions you supply. …","date_published":"2026-11-05T23:03:42.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.15-rc2"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/05/ollama-v0-40-15-1741/","url":"https://labledgerdesk.pages.dev/b/2026/11/05/ollama-v0-40-15-1741/","title":"Ollama v0.40.15","content_text":"MiMo V2.6 Xiaomi MiMo's mimo-v2.6 , a 9B model built for code and agent workflows, is now supported in Ollama. ollama run mimo-v2.6 AutoTrust decision models AutoTrust decision models read text and return choices, yes or no probabilities, and ratings for the questions you supply. …","date_published":"2026-11-05T21:41:21.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.15"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/05/v0-40-15-rc1-llama-cpp-version-bump-b12111-19530/","url":"https://labledgerdesk.pages.dev/b/2026/11/05/v0-40-15-rc1-llama-cpp-version-bump-b12111-19530/","title":"v0.40.15-rc1: llama.cpp: version bump b12111 (#19530)","content_text":"What's Changed mlx: load bundled CUDA libraries in-process on Linux by @dhiltgen in #19433 model: add MiMo V2.6 model support by @dhiltgen in #19400 Full Changelog : v0.40.14...v0.40.15-rc0","date_published":"2026-11-05T01:31:38.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.15-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/03/ollama-v0-40-15/","url":"https://labledgerdesk.pages.dev/b/2026/11/03/ollama-v0-40-15/","title":"Ollama v0.40.15","content_text":"What's Changed mlx: load bundled CUDA libraries in-process on Linux by @dhiltgen in #19433 model: add MiMo V2.6 model support by @dhiltgen in #19400 Full Changelog : v0.40.14...v0.40.15-rc0","date_published":"2026-11-03T22:52:31.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.15-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/03/ollama-v0-40-14/","url":"https://labledgerdesk.pages.dev/b/2026/11/03/ollama-v0-40-14/","title":"Ollama v0.40.14","content_text":"What's Changed ci: pin buildx to workaround deadlock by @dhiltgen in #19373 openai: preserve readable Responses input content by @hoyyeva in #19192 mlx: sort long top-k rows in chunks by @dhiltgen in #19392 app: make ChatGPT disconnect failures retryable by @hoyyeva in #19336 launch: use model output limits for…","date_published":"2026-11-03T01:40:40.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.14-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/11/03/ollama-v0-40-14-457/","url":"https://labledgerdesk.pages.dev/b/2026/11/03/ollama-v0-40-14-457/","title":"Ollama v0.40.14","content_text":"What's Changed ollama launch claude now applies the model's output token limit for Claude Code, so cloud models no longer run into the 128k default cap Cloud web search responses on /api/chat are now decoded correctly, including compressed responses OpenAI-compatible endpoints no longer override a model's packaged…","date_published":"2026-11-03T00:59:50.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.14"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/28/v0-40-13-rc1-llama-cpp-workaround-metal-unary-crash-19372/","url":"https://labledgerdesk.pages.dev/b/2026/10/28/v0-40-13-rc1-llama-cpp-workaround-metal-unary-crash-19372/","title":"v0.40.13-rc1: llama.cpp: workaround Metal unary crash (#19372)","content_text":"This temporary patch works around a regression on older versions of MacOS ggml_metal_library_compile_all: failed to build 'unary' library","date_published":"2026-10-28T22:20:04.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.13-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/28/v0-40-13-llama-cpp-workaround-metal-unary-crash-19372/","url":"https://labledgerdesk.pages.dev/b/2026/10/28/v0-40-13-llama-cpp-workaround-metal-unary-crash-19372/","title":"Ollama v0.40.13","content_text":"LFM3 on Linux and Windows Liquid AI's LFM3 models, including text-to-speech in lfm3-audio and lfm3-omni , now run on Linux and Windows as well as Apple silicon. Full Changelog : v0.40.12...v0.40.13","date_published":"2026-10-28T22:20:04.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.13"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/28/ollama-v0-40-13/","url":"https://labledgerdesk.pages.dev/b/2026/10/28/ollama-v0-40-13/","title":"Ollama v0.40.13","content_text":"What's Changed llm: LFM3 GGUF fixes and llama-server speech by @dhiltgen in #19344 llama.cpp: version bump b11963 by @dhiltgen in #19360 Full Changelog : v0.40.12...v0.40.13-rc0","date_published":"2026-10-28T19:55:04.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.13-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/27/ollama-v0-40-12-182/","url":"https://labledgerdesk.pages.dev/b/2026/10/27/ollama-v0-40-12-182/","title":"Ollama v0.40.12","content_text":"Liquid AI LFM3 LFM3 is Liquid AI's new family of small models built for local agents, tool calling, and structured output, on Liquid's new tri-hybrid architecture. …","date_published":"2026-10-27T16:29:19.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.12-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/27/ollama-v0-40-12-183/","url":"https://labledgerdesk.pages.dev/b/2026/10/27/ollama-v0-40-12-183/","title":"Ollama v0.40.12","content_text":"Liquid AI LFM3 LFM3 is Liquid AI's new family of small models built for local agents, tool calling, and structured output, on Liquid's new tri-hybrid architecture. …","date_published":"2026-10-27T15:47:18.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.12"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/27/ollama-v0-40-12/","url":"https://labledgerdesk.pages.dev/b/2026/10/27/ollama-v0-40-12/","title":"Ollama v0.40.12","content_text":"What's Changed mlx: preserve small-image detail for compact Gemma 4 vision towers by @dhiltgen in #19247 MLX, MLX-C: version bump - prompt speed up for gemma/flint/laguna by @dhiltgen in #19291 mlx: add DFlash2 support by @dhiltgen in #19149 llama.cpp: version bump b11901 by @dhiltgen in #19311 Full Changelog …","date_published":"2026-10-27T00:07:12.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.12-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/23/v0-40-11-rc0-voice-add-voice-catalog-for-phonon-19242/","url":"https://labledgerdesk.pages.dev/b/2026/10/23/v0-40-11-rc0-voice-add-voice-catalog-for-phonon-19242/","title":"v0.40.11-rc0: voice: add voice catalog for Phonon (#19242)","content_text":"What's Changed qwen3coder: preserve literal function closing tags in parameters by @hoyyeva in #19228 launch: use allocated context for OpenCode by @hoyyeva in #19193 decision: allow empty state for vllm-d3 by @SynapShift in #19173 llama.cpp: version bump b11845 by @dhiltgen in #19250 mlxrunner: spell out symbols…","date_published":"2026-10-23T23:15:08.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.11-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/23/v0-40-11-voice-add-voice-catalog-for-phonon-19242/","url":"https://labledgerdesk.pages.dev/b/2026/10/23/v0-40-11-voice-add-voice-catalog-for-phonon-19242/","title":"Ollama v0.40.11","content_text":"Speech Ollama can now generate speech on macOS. ollama run plays text or a chat model's reply aloud as the audio is generated, and apps can stream audio through the OpenAI-compatible /v1/audio/speech endpoint. ollama run phonon \" Hello from Ollama. …","date_published":"2026-10-23T23:15:08.000Z","authors":[{"name":"Ollama"}],"tags":["note"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.11"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/23/v0-40-10-rc0-update-gemma-renderers-19241/","url":"https://labledgerdesk.pages.dev/b/2026/10/23/v0-40-10-rc0-update-gemma-renderers-19241/","title":"v0.40.10-rc0: Update gemma renderers (#19241)","content_text":"renderers: add gemma4-nemotron renderer Add a Gemma 4 renderer variant that renders assistant content before its tool calls rather than after the tool responses. Models opt in with RENDERER gemma4-nemotron; gemma4 auto-detection is unchanged. …","date_published":"2026-10-23T14:12:11.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.10-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/23/v0-40-10-update-gemma-renderers-19241/","url":"https://labledgerdesk.pages.dev/b/2026/10/23/v0-40-10-update-gemma-renderers-19241/","title":"Ollama v0.40.10","content_text":"NVIDIA Nemotron 3.5 Bolt E4B Nemotron 3.5 Bolt E4B is NVIDIA’s smaller, Gemma 4-based vision-language model for fast local conversations, image understanding, and tool use. It supports a 128K-token context window. The model includes a DFlash draft for speculative decoding. …","date_published":"2026-10-23T14:12:11.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.10"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/22/v0-40-9-rc3-support-ga-gradium-phonon-checkpoints-19227/","url":"https://labledgerdesk.pages.dev/b/2026/10/22/v0-40-9-rc3-support-ga-gradium-phonon-checkpoints-19227/","title":"v0.40.9-rc3: Support GA Gradium Phonon checkpoints (#19227)","content_text":"Replace the prerelease checkpoint path with Gradium's GA format, including default-voice import and speech conditioning. Set the 0.40.9 minimum, update the speech docs, and test import, inference setup, and rejection of the old checkpoint.","date_published":"2026-10-22T22:51:43.000Z","authors":[{"name":"Ollama"}],"tags":["note"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.9-rc3"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/22/v0-40-9-support-ga-gradium-phonon-checkpoints-19227/","url":"https://labledgerdesk.pages.dev/b/2026/10/22/v0-40-9-support-ga-gradium-phonon-checkpoints-19227/","title":"Ollama v0.40.9","content_text":"Flint on Linux and Windows Flint now runs on Linux and Windows as well as macOS. Other improvements Streaming OpenAI Responses and Anthropic Messages requests now report generation failures instead of ending silently after partial output. …","date_published":"2026-10-22T22:51:43.000Z","authors":[{"name":"Ollama"}],"tags":["launch"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.9"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/22/v0-40-9-rc2-flint-fix-replies-with-thinking-disabled-19216/","url":"https://labledgerdesk.pages.dev/b/2026/10/22/v0-40-9-rc2-flint-fix-replies-with-thinking-disabled-19216/","title":"v0.40.9-rc2: flint: fix replies with thinking disabled (#19216)","content_text":"flint: fix replies with thinking disabled Close the thinking section when think:false so Flint answers directly. Default and think:true prompts are unchanged. update template","date_published":"2026-10-22T18:38:27.000Z","authors":[{"name":"Ollama"}],"tags":["note"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.9-rc2"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/22/ollama-v0-40-9/","url":"https://labledgerdesk.pages.dev/b/2026/10/22/ollama-v0-40-9/","title":"Ollama v0.40.9","content_text":"Ollama published release v0.40.9 on its official releases feed. The feed did not include notes, so this brief is the version and the primary source, nothing more.","date_published":"2026-10-22T01:33:10.000Z","authors":[{"name":"Ollama"}],"tags":["launch"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.9-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/21/v0-40-9-rc0-ci-switch-to-released-woa-ctk-19185/","url":"https://labledgerdesk.pages.dev/b/2026/10/21/v0-40-9-rc0-ci-switch-to-released-woa-ctk-19185/","title":"v0.40.9-rc0: ci: switch to released WoA CTK (#19185)","content_text":"We were still using a pre-release, and the download site is flaky, so lets get current on the latest CUDA Toolkit for WoA support.","date_published":"2026-10-21T20:12:06.000Z","authors":[{"name":"Ollama"}],"tags":["launch","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.9-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/21/v0-40-8-rc0-mlx-add-flint-model-support-19181/","url":"https://labledgerdesk.pages.dev/b/2026/10/21/v0-40-8-rc0-mlx-add-flint-model-support-19181/","title":"v0.40.8-rc0: mlx: add Flint model support (#19181)","content_text":"Preserve source tool JSON alongside decoded API values so Flint can retain numeric types and schema order in prompts. Existing API marshaling is unchanged. Assemble ModelOpt expert banks during safetensors import, and reject direct Flint NVFP4 creation before conversion. …","date_published":"2026-10-21T14:45:50.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.8-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/21/v0-40-8-mlx-add-flint-model-support-19181/","url":"https://labledgerdesk.pages.dev/b/2026/10/21/v0-40-8-mlx-add-flint-model-support-19181/","title":"Ollama v0.40.8","content_text":"Recursive Flint Flint brings Recursive's coding model to Ollama. It is designed for the work coding agents do across a repository: understanding unfamiliar code, tracing bugs, implementing features, and adding tests. …","date_published":"2026-10-21T14:45:50.000Z","authors":[{"name":"Ollama"}],"tags":["launch","hardware","open"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.8"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/20/v0-40-7-rc1-mlx-deduplicate-cuda-runtime-payloads-17956/","url":"https://labledgerdesk.pages.dev/b/2026/10/20/v0-40-7-rc1-mlx-deduplicate-cuda-runtime-payloads-17956/","title":"v0.40.7-rc1: mlx: deduplicate CUDA runtime payloads (#17956)","content_text":"Deduplicate MLX CUDA runtime payloads Move CUDA libraries shared by the GGML and MLX backends into the sibling CUDA payload, and teach in-process and subprocess MLX loading to resolve that packaged dependency directory. …","date_published":"2026-10-20T21:15:15.000Z","authors":[{"name":"Ollama"}],"tags":["launch","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.7-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/20/v0-40-7-mlx-deduplicate-cuda-runtime-payloads-17956/","url":"https://labledgerdesk.pages.dev/b/2026/10/20/v0-40-7-mlx-deduplicate-cuda-runtime-payloads-17956/","title":"Ollama v0.40.7","content_text":"What's Changed pplx-decider is a 27B decision model from Perplexity, fine-tuned from Qwen3.8-27B Continuing an OpenAI-compatible Responses conversation no longer fails when switching between models with and without thinking, or when the history contains another provider's compaction items Fixed streaming Responses…","date_published":"2026-10-20T21:15:15.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.7"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/17/v0-40-7-rc0-mlx-add-pplx-decider-v1-2-support-19072/","url":"https://labledgerdesk.pages.dev/b/2026/10/17/v0-40-7-rc0-mlx-add-pplx-decider-v1-2-support-19072/","title":"v0.40.7-rc0: mlx: add PPLX Decider v1.2 support (#19072)","content_text":"Implement the Qwen3.5-backed decision scorer with noncausal attention, the publisher readout, image handling, and System One prompt encoding.","date_published":"2026-10-17T21:45:55.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.7-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/16/v0-40-6-docs-add-model-grouping-to-cloud-usage-19069/","url":"https://labledgerdesk.pages.dev/b/2026/10/16/v0-40-6-docs-add-model-grouping-to-cloud-usage-19069/","title":"Ollama v0.40.6","content_text":"What's Changed win: prevent orphaned runners by @dhiltgen in #18500 gguf: add the q2_0 tensor and file types by @jmorganca in #18967 Launch clef in Ollama's cloud by @jberg5 in #19036 use identifiers for --ask mode question keys by @jberg5 in #19041 Full Changelog : v0.40.5...v0.40.6-rc0","date_published":"2026-10-16T23:01:01.000Z","authors":[{"name":"Ollama"}],"tags":["launch"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.6"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/16/v0-40-6-rc1-transfer-support-proxies-on-registry-redirect-targets-19065/","url":"https://labledgerdesk.pages.dev/b/2026/10/16/v0-40-6-rc1-transfer-support-proxies-on-registry-redirect-targets-19065/","title":"v0.40.6-rc1: transfer: support proxies on registry redirect targets (#19065)","content_text":"Behind a proxy, defer redirect-target DNS to the proxy; without one, keep the resolve-and-pin path. Trust default-registry/CDN hostnames with no local DNS. Fix a nil CancelFunc crash in blobDownload.release.","date_published":"2026-10-16T22:00:17.000Z","authors":[{"name":"Ollama"}],"tags":["launch"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.6-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/16/v0-40-6-rc0-merge-pull-request-19041-from-ollama-jberg-clef-question-ids/","url":"https://labledgerdesk.pages.dev/b/2026/10/16/v0-40-6-rc0-merge-pull-request-19041-from-ollama-jberg-clef-question-ids/","title":"v0.40.6-rc0: Merge pull request #19041 from ollama/jberg/clef-question-ids","content_text":"What's Changed win: prevent orphaned runners by @dhiltgen in #18500 gguf: add the q2_0 tensor and file types by @jmorganca in #18967 Launch clef in Ollama's cloud by @jberg5 in #19036 use identifiers for --ask mode question keys by @jberg5 in #19041 Full Changelog : v0.40.5...v0.40.6-rc0","date_published":"2026-10-16T01:09:41.000Z","authors":[{"name":"Ollama"}],"tags":["launch"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.6-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/15/ollama-v0-40-5/","url":"https://labledgerdesk.pages.dev/b/2026/10/15/ollama-v0-40-5/","title":"Ollama v0.40.5","content_text":"NVIDIA Nemotron 3.5 Super NVIDIA Nemotron 3.5 Super is a vision-language model for coding, reasoning, tool use, and document understanding. …","date_published":"2026-10-15T17:03:18.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.5-rc2"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/15/ollama-v0-40-5-123/","url":"https://labledgerdesk.pages.dev/b/2026/10/15/ollama-v0-40-5-123/","title":"Ollama v0.40.5","content_text":"NVIDIA Nemotron 3.5 Super NVIDIA Nemotron 3.5 Super is a vision-language model for coding, reasoning, tool use, and document understanding. …","date_published":"2026-10-15T16:11:12.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.5"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/15/v0-40-5-rc1-new-model-19030/","url":"https://labledgerdesk.pages.dev/b/2026/10/15/v0-40-5-rc1-new-model-19030/","title":"v0.40.5-rc1: New model... (#19030)","content_text":"ci: shorten Windows Vulkan build path Keep the nested shader-generator compiler probes below Windows path limits in checkouts with longer repository names. …","date_published":"2026-10-15T14:28:26.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.5-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/14/ollama-v0-40-5/","url":"https://labledgerdesk.pages.dev/b/2026/10/14/ollama-v0-40-5/","title":"Ollama v0.40.5","content_text":"What's Changed mlx: port the Metal custom kernels to CUDA by @dhiltgen in #18657 docs: model recs and language standardization by @anra8571 in #18981 openai: accept custom tool calls in Responses history by @hoyyeva in #18911 app: clarify cloud payment errors by @willietran in #18986 MLX: version bump by @dhiltgen in…","date_published":"2026-10-14T19:45:22.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.5-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/13/ollama-v0-40-4/","url":"https://labledgerdesk.pages.dev/b/2026/10/13/ollama-v0-40-4/","title":"Ollama v0.40.4","content_text":"What's Changed llama: update llama.cpp to b11521 for decision models and EmbeddingGemma 2 by @jmorganca in #18917 server: reuse verified manifests and loaded models in inference routes by @dhiltgen in #18947 mlx: compact restored recurrent state after speculative rollback by @dhiltgen in #18955 launch: include…","date_published":"2026-10-13T22:03:22.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.4-rc2"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/13/ollama-v0-40-4-106/","url":"https://labledgerdesk.pages.dev/b/2026/10/13/ollama-v0-40-4-106/","title":"Ollama v0.40.4","content_text":"Decision models in the CLI Decision models can now be used directly from the terminal with ollama run — ask yes/no questions, classify text into labels, or score against a rubric: ollama run nimble --ask \"Is a refund being requested?\"","date_published":"2026-10-13T21:23:05.000Z","authors":[{"name":"Ollama"}],"tags":["note"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.4"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/13/v0-40-4-rc1-add-kev-decision-scoring-to-mlx-18976/","url":"https://labledgerdesk.pages.dev/b/2026/10/13/v0-40-4-rc1-add-kev-decision-scoring-to-mlx-18976/","title":"v0.40.4-rc1: Add Kev decision scoring to MLX (#18976)","content_text":"Add Kev decision scoring to MLX Load Kev models with their FP32 decision head and release metadata. Support System One scoring with shared-state caching and Kev-specific limits. address comments","date_published":"2026-10-13T04:50:41.000Z","authors":[{"name":"Ollama"}],"tags":["launch","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.4-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/12/ollama-v0-40-4/","url":"https://labledgerdesk.pages.dev/b/2026/10/12/ollama-v0-40-4/","title":"Ollama v0.40.4","content_text":"What's Changed llama: update llama.cpp to b11521 for decision models and EmbeddingGemma 2 by @jmorganca in #18917 server: reuse verified manifests and loaded models in inference routes by @dhiltgen in #18947 mlx: compact restored recurrent state after speculative rollback by @dhiltgen in #18955 launch: include…","date_published":"2026-10-12T19:00:12.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.4-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/11/ollama-v0-40-3/","url":"https://labledgerdesk.pages.dev/b/2026/10/11/ollama-v0-40-3/","title":"Ollama v0.40.3","content_text":"What's Changed ollama launch claude no longer shows a warning about Claude connectors by @hoyyeva in #18878 ollama launch codex no longer warns about service tiers by @hoyyeva in #18879 Paused background model upgrades, which added overhead to every request in high-volume use such as embeddings, by @jmorganca in…","date_published":"2026-10-11T01:40:01.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.3-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/09/ollama-v0-40-3/","url":"https://labledgerdesk.pages.dev/b/2026/10/09/ollama-v0-40-3/","title":"Ollama v0.40.3","content_text":"What's Changed ollama launch claude no longer shows a warning about Claude connectors by @hoyyeva in #18878 ollama launch codex no longer warns about service tiers by @hoyyeva in #18879 Paused background model upgrade for embedding and other request performance, by @jmorganca in #18908 Full Changelog …","date_published":"2026-10-09T17:36:11.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.3"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/08/ollama-v0-40-2/","url":"https://labledgerdesk.pages.dev/b/2026/10/08/ollama-v0-40-2/","title":"Ollama v0.40.2","content_text":"Model upgrades Models downloaded with earlier versions of Ollama are upgraded in the background the first time you run them, for better performance and compatibility when running on llama.cpp. To make downgrading safe, Ollama keeps the original copy as a backup, so upgraded models are temporarily kept on disk. …","date_published":"2026-10-08T17:44:45.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.2"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/08/v0-40-2-rc0-server-hide-duplicate-and-downgrade-guards-from-list-18874/","url":"https://labledgerdesk.pages.dev/b/2026/10/08/v0-40-2-rc0-server-hide-duplicate-and-downgrade-guards-from-list-18874/","title":"v0.40.2-rc0: server: hide duplicate and downgrade guards from list (#18874)","content_text":"When a legacy GGUF which requires our llama.cpp patch set is loaded, we do a lazy conversion on disk to make it llama.cpp compatible. To support downgrades, we temporarily hold both the new and old GGUFs and create a shadow v1 manifest tag to prevent the downgraded server from deleting the new blobs. …","date_published":"2026-10-08T15:54:06.000Z","authors":[{"name":"Ollama"}],"tags":["launch","llm"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.2-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/07/ollama-v0-40-1/","url":"https://labledgerdesk.pages.dev/b/2026/10/07/ollama-v0-40-1/","title":"Ollama v0.40.1","content_text":"What's Changed server: proxy cloud usage and balance APIs by @drifkin in #18829 llama: fix clef head reads past 2GiB on windows by @Gigrise in #18777 cmd: remove account step from CLI onboarding by @hoyyeva in #18826 manifest: avoid symlinks on Windows by @dhiltgen in #18852 docs: fix 6 dead links in README community…","date_published":"2026-10-07T23:22:59.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.1-rc0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/07/ollama-v0-40-1-84/","url":"https://labledgerdesk.pages.dev/b/2026/10/07/ollama-v0-40-1-84/","title":"Ollama v0.40.1","content_text":"What's Changed server: proxy cloud usage and balance APIs by @drifkin in #18829 llama: fix clef head reads past 2GiB on windows by @Gigrise in #18777 cmd: remove account step from CLI onboarding by @hoyyeva in #18826 manifest: avoid symlinks on Windows by @dhiltgen in #18852 docs: fix 6 dead links in README community…","date_published":"2026-10-07T22:22:54.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/06/ollama-v0-40-0-76/","url":"https://labledgerdesk.pages.dev/b/2026/10/06/ollama-v0-40-0-76/","title":"Ollama v0.40.0","content_text":"What's Changed Models run on MLX on Apple Silicon by default In this release, on Apple Silicon devices, model architectures supported by the MLX runtime will automatically run on MLX. …","date_published":"2026-10-06T17:46:39.000Z","authors":[{"name":"Ollama"}],"tags":["launch","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.0"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/06/v0-40-0-rc6-model-add-multimodal-embeddings-18820/","url":"https://labledgerdesk.pages.dev/b/2026/10/06/v0-40-0-rc6-model-add-multimodal-embeddings-18820/","title":"v0.40.0-rc6: model: add multimodal embeddings (#18820)","content_text":"Implements the EmbeddingGemma2Model architecture on the MLX runner: 24-layer bidirectional text encoder with PLE, shared gemma4 vision/audio towers, mean-pool + L2 output. /api/embed accepts per-item media via input dicts.","date_published":"2026-10-06T15:08:38.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.0-rc6"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/06/ollama-v0-40-0/","url":"https://labledgerdesk.pages.dev/b/2026/10/06/ollama-v0-40-0/","title":"Ollama v0.40.0","content_text":"What's Changed Models run on MLX on Apple Silicon by default In this release, on Apple Silicon devices, model architectures supported by the MLX runtime will automatically run on MLX. …","date_published":"2026-10-06T01:53:06.000Z","authors":[{"name":"Ollama"}],"tags":["launch","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.0-rc5"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/06/v0-40-0-rc4-mlx-version-bump-18720/","url":"https://labledgerdesk.pages.dev/b/2026/10/06/v0-40-0-rc4-mlx-version-bump-18720/","title":"v0.40.0-rc4: MLX: version bump (#18720)","content_text":"MLX: version bump add scopes for unit tests to reduce memory usage","date_published":"2026-10-06T00:41:51.000Z","authors":[{"name":"Ollama"}],"tags":["note","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.0-rc4"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/04/v0-40-0-rc3-pull-allow-rcs-to-pull-matching-min-version-18790/","url":"https://labledgerdesk.pages.dev/b/2026/10/04/v0-40-0-rc3-pull-allow-rcs-to-pull-matching-min-version-18790/","title":"v0.40.0-rc3: pull: allow RCs to pull matching min_version (#18790)","content_text":"Strip off pre-release from the version string so RCs can pull models for the same release.","date_published":"2026-10-04T23:15:37.000Z","authors":[{"name":"Ollama"}],"tags":["launch"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.0-rc3"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/04/ollama-v0-40-0/","url":"https://labledgerdesk.pages.dev/b/2026/10/04/ollama-v0-40-0/","title":"Ollama v0.40.0","content_text":"What's Changed Models run on MLX on Apple Silicon by default In this release, on Apple Silicon devices, model architectures supported by the MLX runtime will automatically run on MLX. …","date_published":"2026-10-04T16:37:17.000Z","authors":[{"name":"Ollama"}],"tags":["launch","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.0-rc2"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/04/v0-40-0-rc1-mlx-match-publisher-tokenizer-semantics-18779/","url":"https://labledgerdesk.pages.dev/b/2026/10/04/v0-40-0-rc1-mlx-match-publisher-tokenizer-semantics-18779/","title":"v0.40.0-rc1: mlx: match publisher tokenizer semantics (#18779)","content_text":"mlx: match publisher tokenizer semantics Honor pretokenizer stage order, split behavior, Unicode boundaries, added-token normalization, and ranked BPE merges. Handle empty added tokens and empty Metaspace input consistently. …","date_published":"2026-10-04T14:45:15.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.40.0-rc1"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/02/ollama-v0-35-1/","url":"https://labledgerdesk.pages.dev/b/2026/10/02/ollama-v0-35-1/","title":"Ollama v0.35.1","content_text":"What's Changed feat: allow ten web searches per response by @ParthSareen in #18602 MLX: version bump by @dhiltgen in #18651 llama.cpp: version bump b11232 by @dhiltgen in #18652 create: support explicit model capabilities by @dhiltgen in #18708 Full Changelog : v0.35.0...v0.35.1-rc0","date_published":"2026-10-02T03:23:32.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.35.1-rc2"},{"id":"https://labledgerdesk.pages.dev/b/2026/10/02/ollama-v0-35-1-60/","url":"https://labledgerdesk.pages.dev/b/2026/10/02/ollama-v0-35-1-60/","title":"Ollama v0.35.1","content_text":"Clef decision models Ollama now supports Clef and Clef Flash , Cloudflare's new open-source decision models, through /v1/systemone . Clef (27B) and Clef Flash (9B) are multimodal: requests can now include images alongside the text state, shared by all questions and scored jointly with it. …","date_published":"2026-10-02T01:05:01.000Z","authors":[{"name":"Ollama"}],"tags":["note","llm","hardware"],"external_url":"https://github.com/ollama/ollama/releases/tag/v0.35.1"}]}