Commit Graph

  • c330ea33ed qwen3next: handle mixed recurrent batches jmorganca/qwen3-concurrent jmorganca 2026-02-05 11:47:27 -08:00
  • c61023f554 ollamarunner: Fix off by one error with numPredict v0.15.5-rc4 Jesse Gross 2026-02-04 15:36:11 -08:00
  • d25535c3f3 qwen3next: avoid inplace sigmoid for shared gate (#14077) Jeffrey Morgan 2026-02-04 15:50:02 -08:00
  • c323161f24 cmd: helpful error message for remote models (#14057) Bruce MacDonald 2026-02-04 14:55:11 -08:00
  • 255579aaa7 qwen3next: fix issue in delta net (#14075) v0.15.5-rc3 Jeffrey Morgan 2026-02-04 13:40:38 -08:00
  • f7102ba826 runner: discard compute results if sequence replaced mid-batch (#14072) Jeffrey Morgan 2026-02-04 13:19:48 -08:00
  • cefabd79a8 Revert "cmd: claude launch improvements (#14064)" (#14071) Jeffrey Morgan 2026-02-04 09:10:37 -08:00
  • df70249520 server: optimize chatPrompt to reduce tokenization calls (#14040) Jeffrey Morgan 2026-02-04 01:21:31 -08:00
  • 77eb2ca619 model: add qwen3-next architecture (#14051) v0.15.5-rc2 Jeffrey Morgan 2026-02-03 23:27:21 -08:00
  • ee25219edd cmd: claude launch improvements (#14064) Parth Sareen 2026-02-03 22:33:58 -05:00
  • b1fccabb34 Revert "Update vendored llama.cpp to b7847" (#14061) Jeffrey Morgan 2026-02-03 18:39:36 -08:00
  • a6355329bf cmd: open browser on ollama signin when available (#14055) Bruce MacDonald 2026-02-03 16:42:09 -08:00
  • 55746e31fa ggml: add MLA flash attention config for GLM-4.7-flash fix-glm-4.7-flash-mla-config jmorganca 2026-02-03 12:57:48 -08:00
  • 0398b24b42 cmd: launch defaults (#14035) v0.15.5-rc1 Parth Sareen 2026-02-03 02:19:11 -05:00
  • 75b1dddf91 cmd: launch extra params (#14039) Parth Sareen 2026-02-03 02:03:33 -05:00
  • e1e80ffc3e cmd/config: move config location (#14034) Parth Sareen 2026-02-02 22:48:51 -05:00
  • 71896485fd anthropic: add InputTokens to streaming response (#13934) Aleksandr Vukmirovich 2026-02-03 03:29:37 +01:00
  • ef00199fb4 Update vendor ggml code to a5bb8ba4 (#13832) Jeffrey Morgan 2026-02-02 17:31:59 -08:00
  • 152b922265 checkpoint Devon Rifkin 2026-02-02 15:42:57 -08:00
  • 8f4a008139 Add GLM-OCR vision model support (#14024) v0.15.5-rc0 Jeffrey Morgan 2026-02-02 15:39:18 -08:00
  • d8cc798c2b glm 4.7 flash support on experimental engine (#13838) Patrick Devine 2026-02-02 15:22:11 -08:00
  • ba75143e71 WIP aliases Devon Rifkin 2026-02-02 14:59:58 -08:00
  • b202a9b4ce qwen3-coder parser: allow missing opening tool call tag drifkin/qwen3-coder-opening-tag Devon Rifkin 2026-02-02 12:53:45 -08:00
  • 6582f6da5c llm: Make "do load request" error message more informative Richard Lyons 2026-01-30 16:34:21 +01:00
  • 0334ffa625 server: use tiered VRAM-based default context length Jesse Gross 2026-01-27 16:12:17 -08:00
  • d11fbd2c60 server: fix ollama ps showing configured instead of actual context length Jesse Gross 2026-01-27 16:27:55 -08:00
  • 6a7c3f188e openclaw: run onboarding for fresh installs (#14006) v0.15.4 Jeffrey Morgan 2026-02-01 13:46:45 -08:00
  • 427e2c962a docs: add redirect from clawdbot to openclaw (#14004) Jeffrey Morgan 2026-01-31 20:50:42 -08:00
  • 27db7f806f cmd/config: rename integration to openclaw (#13979) v0.15.3 Thanh Nguyen 2026-02-01 06:31:13 +07:00
  • 3590fbfa76 runner: fix typo 'baackend' -> 'backend' in error messages (#13645) Dhiraj Lochib 2026-02-01 02:56:20 +05:30
  • cd0094f772 added stakpak to web & desktop (#13961) noureldin-azzab 2026-01-31 23:04:34 +02:00
  • 06bc8e6712 docs: add Screenpipe to Community Integrations (#13906) Louis Beaumont 2026-01-31 12:49:52 -08:00
  • fc5f9bb448 docs: remove unsupported quantizations (#13982) frob 2026-01-31 21:46:20 +01:00
  • a0740f7ef7 docs: add GB10 to supported devices (#13987) frob 2026-01-31 21:45:27 +01:00
  • a0923cbdd0 cmd: ollama launch add placeholder text for selector (#13966) Parth Sareen 2026-01-29 12:48:49 -05:00
  • f92e362b2e cmd: capitalize Ollama in serve command help text (#13965) Seokrin Taron Sung 2026-01-30 02:47:53 +09:00
  • aa23d8ecd2 docs: update installation command for OpenCode CLI (#13971) Tincho 2026-01-29 14:47:02 -03:00
  • e6f5a982d3 cmd: add usage cmd to chat to see token consumption brucemacd/usage-cli Bruce MacDonald 2026-01-27 17:14:25 -08:00
  • 7b62c41060 cmd/config: use envconfig.Host() for base API in launch config packages (#13937) Gabe Goodhart 2026-01-27 14:30:00 -07:00
  • 26acab64b7 docs: add clawdbot (#13925) Parth Sareen 2026-01-26 21:32:54 -05:00
  • e0f03790b1 parsers/ministral: fix nested tool call parsing by counting brace nesting (#13905) Gyungrai Wang 2026-01-27 08:03:43 +09:00
  • 3ab842b0f5 cmd: clawdbot config fixes (#13922) v0.15.2 Parth Sareen 2026-01-26 17:34:29 -05:00
  • b8e8ef8929 cmd: ollama launch clawdbot (#13921) Parth Sareen 2026-01-26 16:40:59 -05:00
  • 465d124183 cmd: fix opencode config (#13894) v0.15.1 Parth Sareen 2026-01-24 21:42:56 -05:00
  • d310e56fa3 cmd: add fallback for claude (#13892) Parth Sareen 2026-01-24 21:26:01 -05:00
  • a1ca428c90 glm4moelite: fix attention scale calculation (#13893) Jeffrey Morgan 2026-01-24 17:48:09 -08:00
  • 16750865d1 glm4moelite: quantize more tensors to q8_0 and avoid double BOS token (#13891) v0.15.1-rc1 Jeffrey Morgan 2026-01-24 16:33:54 -08:00
  • f3b476c592 build: add -O3 optimization to CGO flags (#13877) v0.15.1-rc0 Jeffrey Morgan 2026-01-24 10:55:38 -08:00
  • 5267d31d56 docs: ollama launch (#13852) Parth Sareen 2026-01-24 02:18:50 -05:00
  • b44f56319f README: Update the "Ollama for ruby" to the most popular and maintained ruby gem. (#13855) Stillhart 2026-01-24 07:24:52 +01:00
  • 0209c268bb llama: fix CUDA MMA errors in release build (#13874) v0.15.0-rc6 v0.15.0 Jeffrey Morgan 2026-01-23 20:10:04 -08:00
  • 8e22b09e2c ggml-cuda: fix fattn build for GLM 4.7 flash support fix-cuda12-fattn-shmem Jeffrey Morgan 2026-01-24 03:12:49 +00:00
  • 912d984346 llama: fix fattn-tile shared memory overflow on sm_50/52 (#13872) v0.15.0-rc5 Jeffrey Morgan 2026-01-23 19:22:32 -08:00
  • aae6ecbaff cmd: rename ollama config to ollama launch (#13871) v0.15.0-rc4 Parth Sareen 2026-01-23 21:40:40 -05:00
  • 64737330a4 Re-apply "model: add MLA absorption for glm4moelite" with fix (#13870) Jeffrey Morgan 2026-01-23 18:40:28 -08:00
  • 2eda97f1c3 Revert "model: add MLA absorption for glm4moelite (#13810)" (#13869) v0.15.0-rc3 Jeffrey Morgan 2026-01-23 17:14:15 -08:00
  • 66831dcf70 x/imagegen: fix image editing support (#13866) v0.15.0-rc2 Jeffrey Morgan 2026-01-23 15:37:17 -08:00
  • 1044b0419a model: add MLA absorption for glm4moelite (#13810) Jeffrey Morgan 2026-01-23 14:47:42 -08:00
  • 771d9280ec cmd: ollama config fix droid model name configuration (#13856) Parth Sareen 2026-01-23 14:44:22 -05:00
  • 862bc0a3bf x/imagegen: respect stream=false in /api/generate (#13853) Jeffrey Morgan 2026-01-22 22:16:39 -08:00
  • c01608b6a1 x/imagegen: add image edit capabilities (#13846) v0.15.0-rc1 Jeffrey Morgan 2026-01-22 20:35:08 -08:00
  • 199c41e16e cmd: ollama config command to help configure integrations to use Ollama (#13712) Parth Sareen 2026-01-22 23:17:11 -05:00
  • 3b3bf6c217 x/imagegen: replace memory estimation with actual weight size (#13848) Jeffrey Morgan 2026-01-22 18:32:41 -08:00
  • f52c21f457 fix: handle Enter key pressed during model loading (#13839) Parth Sareen 2026-01-22 21:32:02 -05:00
  • 8b4410633d Add image generation documentation ollama-imagegen-docs jmorganca 2026-01-22 13:51:56 -08:00
  • c73feaf73d Clean up the manifest and modelpath (#13807) parth/fix-multiline-inputs Patrick Devine 2026-01-21 11:46:17 -08:00
  • cc3ac5fee3 docs: update instructions for ollama config command brucemacd/config-docs Bruce MacDonald 2026-01-21 17:03:41 -08:00
  • b5d0f72f16 x/imagegen: remove qwen_image and qwen_image_edit models (#13827) v0.15.0-rc0 Jeffrey Morgan 2026-01-21 13:37:08 -08:00
  • 148a1be0a3 Clean up the manifest and modelpath (#13807) Patrick Devine 2026-01-21 11:46:17 -08:00
  • e4d0a8f016 draft: model manifest file interface mxyng/model-files Michael Yang 2026-01-19 13:34:24 -08:00
  • 268c2a1df1 fix: remove multiline option in non-experimental mode sunyongyue 2026-01-22 02:02:54 +08:00
  • 110b072200 chore: simplify runner.Execute mxyng/simple-execute Michael Yang 2026-01-19 14:00:05 -08:00
  • d6dd430abd x/imagegen: respect OLLAMA_MODELS for manifests and blobs (#13797) v0.14.3 next-n 2026-01-21 03:31:52 +06:30
  • ae78112c50 test: add lfm2.5-thinking coverage (#13802) Daniel Hiltgen 2026-01-20 12:57:02 -08:00
  • 91a3b77553 x/imagegen: respect OLLAMA_MODELS environment variable fix-imagegen-ollama-models jmorganca 2026-01-20 12:46:20 -08:00
  • 01cf7445f3 model: add lfm2 architecture and LFM2.5-1.2B-Thinking support (#13792) v0.14.3-rc3 Jeffrey Morgan 2026-01-20 12:20:53 -08:00
  • 9ef2106b47 cmd: create blob in parallel with checksum mxyng/async-upload Michael Yang 2026-01-20 09:53:56 -08:00
  • 31085d5e53 fix: use api.GenerateRequest for image generation test (#13793) Jeffrey Morgan 2026-01-20 03:23:31 -08:00
  • 20776ced38 x/imagegen: skip eval when no dtype conversion needed jmorganca/lazy-no-dtype-changes jmorganca 2026-01-19 16:25:37 -08:00
  • 5dc20e91d0 simplify imagegen-auto-detect-create jmorganca 2026-01-19 16:15:15 -08:00
  • c42e9d244f test: add image gen test case (#13698) Daniel Hiltgen 2026-01-19 16:01:31 -08:00
  • e98b5e8b4e /api/show: default to empty model_info (#13785) Devon Rifkin 2026-01-19 15:26:17 -08:00
  • bda8cb7403 cmd: auto-detect image generation models during create jmorganca 2026-01-19 14:43:14 -08:00
  • 68e00c7c36 fix: prevent image generation models from loading during deletion (#13781) v0.14.3-rc2 Jeffrey Morgan 2026-01-19 12:48:34 -08:00
  • 4f138a1749 model: add Glm4MoeLiteForCausalLM architecture to support GLM-4.7-Flash (#13779) Jeffrey Morgan 2026-01-19 12:47:17 -08:00
  • 03bf241c33 x/imagegen: add FP4 quantization support for image generation models (#13773) Jeffrey Morgan 2026-01-19 00:54:54 -08:00
  • a887406c24 x/imagegen: add preliminary support for FLUX.2-klein model (#13772) Jeffrey Morgan 2026-01-18 22:30:49 -08:00
  • d51e95ba7e server: prevent image generation models from reloading on every request (#13771) Jeffrey Morgan 2026-01-18 20:50:04 -08:00
  • 6b2abfb433 server: add tests and fix isHuggingFaceURL edge case parth/decrease-concurrent-download-hf Parth Sareen 2026-01-18 16:45:17 -08:00
  • 805ed4644c server: reduce download concurrency for HuggingFace URLs Parth Sareen 2026-01-18 16:38:49 -08:00
  • 582d93ab22 fix: lazy init MLX for quantization and improve library discovery fix-mlx-quantize-init jmorganca 2026-01-17 22:25:31 -08:00
  • 3d01f2aa34 parsers: refactor Nemotron parser to reuse Qwen3Coder for tool calls (#13764) Jeffrey Morgan 2026-01-17 18:28:52 -08:00
  • 634c416645 Add experimental image generation fields to /api/generate (#13753) Jeffrey Morgan 2026-01-17 18:27:41 -08:00
  • b220bfa0b9 x: clean up some experimental mlx backend code jmorganca/x-cleanup jmorganca 2026-01-17 01:27:12 -08:00
  • 57de86cc61 docs: update claude code docs (#13757) Michael 2026-01-16 22:41:34 -08:00
  • 12719b6e87 MLX - dynamic loading of mlx-c (#13735) v0.14.3-rc1 Daniel Hiltgen 2026-01-16 16:34:22 -08:00
  • a077d996e3 Fix create and show commands for experimental models (#13741) Patrick Devine 2026-01-16 14:31:55 -08:00
  • c23d5095de x/imagegen: clean up image generation code (#13725) Jeffrey Morgan 2026-01-16 12:19:25 -08:00
  • 7601f0e93e server: reject unexpected auth hosts (#13738) Bruce MacDonald 2026-01-16 14:10:36 -05:00
  • aad3f03890 app: allow macOS app to terminate during system shutdown (#13737) v0.14.3-rc0 Eva H 2026-01-16 09:05:04 -05:00