← back to report

ollama/ollama

// all events — 2026-09-19

Type AI Score Description Actor Reason Date
commit 1.00 fix(docs): correct typos found during code review (#17579) Commit message contains explicit AI assi 2026-08-31
commit 1.00 cmake: make external compat patches idempotent (#17948) Commit message contains explicit AI assi 2026-08-26
pr 1.00 model/parsers: accept implicit qwen3coder tool openers Explicitly states it was built by an AI 2026-09-18
pr 1.00 responses: support namespace tool calls The description explicitly states that C 2026-05-22
pr 1.00 docs: include thinking in chat message schema Explicitly states that it was prepared w 2026-09-16
pr 1.00 server: fix MLX structured output after thinking The description explicitly states that A 2026-09-15
pr 1.00 server: add experimental prefill cache persistence across ru Explicitly states the PR was developed w 2026-08-23
pr 1.00 template: render tool function and parameters as JSON Explicitly states it was generated with 2026-09-11
pr 0.91 feat: improve CLI navigation, implement atomic manifest writ Emoji-heavy overview and polished generi 2026-06-21
pr 0.84 fix(cli): abort interactive mode with error when stdout is r Highly polished, repetitive verification 2026-06-21
pr 0.64 Add Talos as an `ollama launch` integration Extremely polished exhaustive prose has 2026-08-28
pr 0.28 readme: add deel to Code Editors & Development Polished promotional copy, though produc 2026-09-06
pr 0.24 Add digest and provider_version to /api/chat responses Polished and highly structured, but stil 2026-09-18
pr 0.24 Add digest and provider_version to /api/chat responses Formal wording is present, though the te 2026-09-18
pr 0.24 server: allow parallel requests for qwen35 / qwen35moe now t Polished and highly structured, but grou 2026-07-12
pr 0.18 fix(deps): bump golang.org/x/image to patch WebP decoder DoS Highly polished advisory summary, but co 2026-09-10
pr 0.16 app: refresh Apps layout and command copy feedback Polished summary is somewhat generic but 2026-09-10
pr 0.12 cli: add first-run onboarding shared with the desktop app Polished but concrete PR summary with de 2026-09-16
pr 0.12 readme: add LLMxRay to Observability & Monitoring Somewhat polished promotional wording, b 2026-09-14
pr 0.08 Add LibreDB Studio to Community Integrations Specific product details and team disclo 2026-09-05
pr 0.08 Fix: minicpm5-2b native tool calls never parse Detailed implementation and test summary 2026-09-17
pr 0.08 feat: implement full jsonschema spec Domain-specific implementation details a 2026-09-16
pr 0.08 anthropic: keep an inline system message out of the system b Highly technical, issue-specific explana 2026-09-15
pr 0.06 MSI Based windows install with "curl" style install script. Detailed technical description with spec 2026-02-05
pr 0.06 Recover panics in CreateHandler's model conversion goroutine Specific root-cause analysis and candid 2026-08-07
pr 0.06 openai: finish reasoning before emitting client tool calls Concise regression description includes 2026-09-12
pr 0.06 docs, scripts: Prevent useradd(8) from failing on Fedora Sil Specific platform details and a visible 2025-09-30
pr 0.06 linux: fix Vulkan inference support on ARM64 Detailed ARM64 and Vulkan-specific conte 2026-09-15
pr 0.06 launch: add Atomic Agent integration Highly specific integration details and 2026-08-25
pr 0.05 llama-server: prepare to remove compatibility patch with man Dense, highly specific technical termino 2026-06-07
pr 0.05 server: serve Prometheus metrics at /metrics when OLLAMA_MET Detailed issue comparison and project-sp 2026-09-17
pr 0.05 app: clarify Claude connection dialog Concrete UI changes and test results are 2026-09-16
commit 0.04 nemotron_h: add MLX vision support (#17714) Detailed domain-specific changes and rev 2026-09-17
commit 0.04 ci: harden release artifact uploads (#18516) Specific release sequencing, retry timin 2026-09-17
commit 0.04 create: add server-side MLX imports and drop GGUF conversion Detailed implementation summary with con 2026-09-15
pr 0.04 windows: use install.ps1 for app upgrades Specific Windows updater details and dep 2026-06-25
pr 0.04 windows: add install.ps1 cache modes Concrete implementation and testing deta 2026-06-25
pr 0.04 api: expose model thinking levels and defaults Detailed implementation behavior and cli 2026-09-15
pr 0.04 app: restore Apps list layout Detailed validation notes and repository 2026-09-17
pr 0.04 MLX: make Qwen3.8 Flash Next memory efficient Highly specific MLX and memory measureme 2026-08-27
pr 0.04 progress: serialize concurrent shutdown writes Specific race details, test commands, an 2026-06-01
pr 0.04 fix(server): parse reasoning on /api/generate even when thin Contains detailed code-path investigatio 2026-09-17
pr 0.04 win: prevent orphaned runners Specific Windows process-management rati 2026-09-17
pr 0.04 openai: reject previous_response_id instead of answering wit Detailed, issue-specific explanation and 2026-09-14
commit 0.03 x/transfer, server: tighten redirect handling for registry r Uses precise security and networking ter 2026-09-17
commit 0.03 MLX, MLX-C: version bump (#18449) Specific version and model details indic 2026-09-15
commit 0.03 llama.cpp: version bump b10729 (#18160) Terse, repository-specific changelog wit 2026-09-01
commit 0.03 Honor model generation defaults (#16471) Specific behavior, metadata sources, and 2026-09-01
pr 0.03 cmd/model: add offline model diff tool Dense repository-specific technical deta 2026-09-02
pr 0.03 app: support preview update channel Concise release-channel description is o 2026-09-18
pr 0.03 tokenizer: keep Unicode punctuation intact in WordPiece Specific Unicode, test, platform, and de 2026-09-15
pr 0.03 test: harden integration tests, lint, add create scope Compact list of concrete test changes an 2026-07-27
pr 0.03 bench: use HumanEval patch prompts Concise, domain-specific benchmark detai 2026-07-30
pr 0.03 nemotron_h: add MLX vision support Dense domain terminology and implementat 2026-08-12
pr 0.03 fix(gemma4): stop dropping tool parameters named like schema Highly specific renderer behavior, templ 2026-09-17
pr 0.03 mlx: budget GPU memory for CUDA devices Terse implementation note with concrete 2026-09-15
pr 0.03 app: add Apps deep link Specific deep-link behavior and validati 2026-09-16
pr 0.03 mlx: deduplicate CUDA runtime payloads Detailed platform-specific packaging ter 2026-08-23
pr 0.03 Move the MLX engine out of x/ Detailed repository structure and specif 2026-09-16
pr 0.03 server: handle thinking in native generate templates Specific issue references and bilingual 2026-09-15
pr 0.03 tools: return buffered output as content when no tool call i Specific Go behavior, issue reference, a 2026-09-15
pr 0.03 cmd/launch: align Claude context suffix with effective cloud Concise changelog with concrete symbols, 2026-09-15
pr 0.03 tools: return buffered output as content when no tool call i Specific Go behavior, issue reference, a 2026-09-15
pr 0.03 gemma4: accept '=' separator in tool call arguments Model-specific parser terminology and pr 2026-08-20
pr 0.03 cmd: honor cancellation while waiting for server Standard problem/fix/test structure with 2026-09-15
pr 0.03 server: bound direct URL resolution attempts Concrete regression details and test com 2026-09-14
pr 0.03 chat: always preserve the most recent user message during tr Specific code paths, model behavior, and 2026-08-20
commit 0.02 tests: fix metadata deletion test flake (#18493) Terse, issue-specific wording with concr 2026-09-17
commit 0.02 llama.cpp: version bump b10969 (#18446) Terse, repository-specific changelog wit 2026-09-15
commit 0.02 API: Deprecate typical_p (#18448) Concise technical change using specific 2026-09-15
commit 0.02 MLX: version bump (#18235) Terse version-bump notes and typical rev 2026-09-14
commit 0.02 llm: keep gemma3n projector off the CPU (#18376) Concise, domain-specific wording and con 2026-09-11
commit 0.02 server: extract GGUF metadata and unify capabilities (#17858 Technical changelog includes concrete pa 2026-09-10
commit 0.02 ci: rebuild MLX macOS test payloads the release can't supply Highly specific CI behavior and terse ph 2026-09-03
commit 0.02 Report cached prompt tokens (#17943) Structured changelog with concrete API a 2026-09-02
commit 0.02 launch: respect OLLAMA_HOST when configuring Pi (#18039) Specific technical diagnosis and behavio 2026-08-26
commit 0.02 MLX: Qwen3.8 Flash Next support (#18032) Terse feature note and review reference, 2026-08-26
pr 0.02 create: quantize gemma4 sparse-MoE expert banks uniformly Specific technical diagnosis and measure 2026-08-24
pr 0.02 api: expose projected context length in model details Brief, tentative technical note lacks AI 2026-08-11
pr 0.02 mlx: lazy initialize to avoid crashes Informal wording and a typo strongly sug 2026-05-14
pr 0.02 app: keep closed macOS windows from reopening on activation Concise platform-specific behavior descr 2026-09-17
pr 0.02 Allow 96-char model names so long HF GGUF pulls validate Casual phrasing, exact issue details, an 2026-09-18
pr 0.02 tests: fix metadata deletion test flake Specific test fix and issue reference in 2026-09-16
pr 0.02 UX: reject oversized pulls by default Concise, domain-specific description wit 2026-09-04
pr 0.02 ci: harden release artifact uploads Contains a typo and concrete CI details, 2026-09-17
pr 0.02 x/transfer, server: tighten redirect handling for registry r Uses precise security and networking ter 2026-09-17
pr 0.02 docs: clarify integrations overview and add assistant icons Brief documentation summary with specifi 2026-09-17
pr 0.02 test: split create integration tests out Terse description includes a concrete co 2026-09-15
pr 0.02 mlx: warm up MLX compile paths on load Specific timing behavior plus an informa 2026-05-10
pr 0.02 docs: fix cloud retirements settings link Minimal, concrete change description wit 2026-09-16
pr 0.02 llm: disable MTP backend sampling on Metal Specific benchmarks, domain terminology, 2026-08-25
pr 0.02 tools: return buffered output when no tool call is parsed Specific parser behavior, regression cov 2026-09-15
pr 0.02 server: propagate cloud stream failures Concise template-based summary with spec 2026-09-15
pr 0.02 model/parsers: parse Gemma 4 BEGIN_ARG tool calls Domain-specific parser details, issue re 2026-09-15
pr 0.02 server: abort downstream stream when cloud upstream truncate Detailed implementation scope and test n 2026-09-15
pr 0.02 MLX, MLX-C: version bump Concise release note with specific MLX-C 2026-09-15
pr 0.02 create: add server-side MLX imports and drop GGUF conversion Detailed domain-specific change summary 2026-03-19
pr 0.02 feat(fs/ggml): support Q2_0 GGUF tensors Terse, domain-specific wording strongly 2026-09-14
commit 0.01 api: expose model thinking levels and defaults (#18473) Terse, issue-specific title with no AI-s 2026-09-18
commit 0.01 app: keep closed macOS windows from reopening on activation Concise, domain-specific commit-style wo 2026-09-18
commit 0.01 docs: clarify integrations overview and add assistant icons Brief conventional documentation change 2026-09-17
commit 0.01 cli: add first-run onboarding shared with the desktop app (# Terse conventional feature summary witho 2026-09-17
commit 0.01 app: add Apps deep link (#18497) Short, conventional commit-style feature 2026-09-17
commit 0.01 app: clarify Claude connection dialog (#18498) Terse, issue-specific commit title with 2026-09-16
commit 0.01 docs: fix cloud retirements settings link (#18496) Brief technical commit title is characte 2026-09-16
commit 0.01 docs: refresh getting started guides (#18450) Brief conventional commit message with n 2026-09-15
commit 0.01 mlx: add mlx patch to docker build context (#18440) Terse, repository-specific commit title 2026-09-14
commit 0.01 docs: add ChatGPT Desktop integration (#18377) Brief, conventional documentation commit 2026-09-14
commit 0.01 app: refresh Apps layout and command copy feedback (#18372) Concise, conventional commit title with 2026-09-14
commit 0.01 llm: raise token repeat limit to 100 and return error instea Terse, domain-specific commit message wi 2026-09-10
commit 0.01 openai: finalize responses at the web search limit (#18328) Terse, domain-specific commit title with 2026-09-08
commit 0.01 app: add a first-use ChatGPT connection intro (#18321) Concise conventional commit title; no AI 2026-09-08
commit 0.01 app: align ChatGPT connection status copy with Claude (#1832 Brief technical commit title with natura 2026-09-08
commit 0.01 server/openai: retry compaction after context overflow (#183 Terse domain-specific title and conventi 2026-09-08
commit 0.01 app: open Codex by default for Ollama launches (#18323) Concise technical commit title with repo 2026-09-08
commit 0.01 fix data races in progress and sched (#18319) Specific concurrency details and technic 2026-09-08
commit 0.01 app: preserve Codex configuration across host changes (#1824 Short conventional commit title without 2026-09-08
commit 0.01 openai: use tsc_ prefix for tool search call IDs (#18296) Specific API terminology and concise com 2026-09-08
commit 0.01 app: harden Codex desktop proxy handling (#18244) Terse, domain-specific commit title with 2026-09-05
commit 0.01 openai: preserve images through response compaction (#18245) Terse, specific commit title; no signs o 2026-09-05
commit 0.01 app: route Codex auto review through selected model (#18246) Concise technical commit title with natu 2026-09-05
commit 0.01 ci: bound and cache the macOS MLX payload build (#18240) Concise, specific CI description with na 2026-09-04
commit 0.01 app: add Ollama to ChatGPT Desktop (#18236) Terse, repository-specific commit-style 2026-09-04
commit 0.01 openai: add Codex compaction support (#18224) Terse, domain-specific commit-style summ 2026-09-04
commit 0.01 openai: add client tool search support (#18223) Terse, domain-specific commit-style summ 2026-09-04
commit 0.01 ci: wire up MLX unit tests for PR runs (#17022) Technical CI notes include specific revi 2026-09-01
commit 0.01 MLX, MLX-C: version bump (#18080) Brief version-bump message with informal 2026-09-01
commit 0.01 build: go deps (#18161) Very terse dependency update with a conv 2026-09-01
commit 0.01 proxy: continue requests when the model catalog changes (#18 Terse, issue-linked commit title with no 2026-08-27
commit 0.01 app: restore system dark mode (#18049) Concise technical commit title; no free- 2026-08-27
commit 0.01 app: consume Claude recommendation mappings (#18042) Specific, terse commit title with domain 2026-08-26
commit 0.01 app: prevent Claude model picker scroll jumps (#18041) Concise issue-linked commit title with n 2026-08-26
commit 0.01 linux: fix llama.cpp docker build (#18040) Brief, concrete build-fix explanation us 2026-08-26
commit 0.01 docs: claude docs (#18006) Very terse documentation commit title. 2026-08-26
commit 0.01 docs: document Claude Desktop integration (#18004) Concise, specific documentation commit t 2026-08-26
commit 0.01 proxy: preserve string content during image fallback (#18002 Terse conventional commit with a specifi 2026-08-25
pr 0.01 server: allow registry cross-host redirects among allowliste Brief domain-specific title and human re 2026-09-18
pr 0.01 README: update Msty link to msty.ai Short, concrete documentation update wit 2026-08-26
pr 0.01 cmd: remove built-in agent Terse, informal revert description is ch 2026-09-11
pr 0.01 docs: document thinking control discovery A terse, specific title with no AI-style 2026-09-17
pr 0.01 server: apply structured outputs in a single pass on thinkin Highly specific implementation details a 2026-09-15
pr 0.01 server: fix intermittent "model not found" errors. Awkward wording, duplicated title contex 2026-09-14
pr 0.01 mlxrunner: implement load progress reporting and stall detec Specific MLX loading behavior and an inc 2026-08-18
pr 0.01 mlxrunner: Release freed KV buffers during speculative decod Concrete memory measurements and impleme 2026-09-17
pr 0.01 Update routes_test.go Extremely terse title-only contribution 2026-09-16
pr 0.01 server: apply structured outputs in a single pass on thinkin Highly specific implementation details a 2026-09-15
pr 0.01 llama.cpp: version bump b10969 Short release-note style description wit 2026-09-14
pr 0.01 server: warn when embedding input is truncated Highly specific reproduction, code behav 2026-08-03
pr 0.01 docs: refresh getting started guides Brief, informal commit message with no n 2026-09-15
pr 0.01 API: Deprecate typical_p Direct technical wording and specific pa 2026-09-14
pr 0.01 app/ui: honor cancellation while waiting for server Terse, domain-specific bug summary with 2026-09-15
pr 0.01 app/server: cancel inference log polling Specific resource and cancellation detai 2026-09-15
pr 0.01 x/transfer: bound authentication retries Technical, concrete retry behavior and t 2026-09-15
pr 0.01 mlx: add mlx patch to docker build context Brief follow-up note with issue referenc 2026-09-14
pr 0.01 docs: add ChatGPT Desktop integration Title-only contribution provides no evid 2026-09-10
pr 0.01 MLX: version bump Minimal version-bump description contain 2026-09-04
commit 0.00 mlxrunner: Release freed KV buffers during speculative decod Detailed MLX diagnosis with concrete mea 2026-09-17
commit 0.00 mlxrunner: lay out model by contract, checkpoint and constru Detailed refactoring rationale uses spec 2026-09-16
commit 0.00 mlxrunner: lay out nn one layer kind per file Specific package and interface reference 2026-09-16
commit 0.00 mlxrunner: read model manifests through the manifest package Concrete manifest and loader details out 2026-09-16
commit 0.00 mlxrunner, cmd, server: fold three small packages into their Highly specific package consolidation de 2026-09-16
commit 0.00 mlx, mlxrunner: move the MLX engine out of x/ Detailed path list and build-system refe 2026-09-16
commit 0.00 tokenizer, ml: remove dead code Concise cleanup summary cites concrete r 2026-09-16
commit 0.00 cmd: launch the MLX runner without the engine dispatcher Specific Cobra behavior and command chan 2026-09-16
commit 0.00 cmd: remove built-in agent (#18393) Terse commit title with no AI-style phra 2026-09-11
commit 0.00 mlxrunner: release the buffers weight loading leaves in the Specific allocator, model, and memory fi 2026-09-10
commit 0.00 mlx: drop the empty-handle checks outside the bindings Precise internal lifecycle details and t 2026-09-04
commit 0.00 mlx: scope array lifetimes instead of pinning and sweeping Highly specific implementation and bench 2026-09-04
commit 0.00 mlxrunner: wait for a killed runner to exit before the sched Highly specific concurrency and signal-h 2026-09-09
commit 0.00 mlxrunner: bound MLX loads by system free memory while other Uses detailed Apple silicon and schedule 2026-09-10
commit 0.00 mlxrunner: evict the active conversation's own checkpoints u Specific implementation behavior, model 2026-09-09
commit 0.00 mlxrunner: keep the reused head of a cached edge safe from e Detailed cache and recurrent-layer behav 2026-09-10
commit 0.00 mlxrunner: keep cache boundaries out of non-causal media ite Concise technical explanation with preci 2026-09-09
commit 0.00 mlxrunner: capture whole-state at close for nodes split out References exact snapshot and eviction i 2026-09-09
commit 0.00 llama.cpp: version bump b10864 (#18317) Minimal conventional version-bump commit 2026-09-10
commit 0.00 app: fix ChatGPT model selector spacing (#18347) Terse, domain-specific commit message wi 2026-09-09
commit 0.00 openai: support standalone named function outputs (#18348) Concise technical commit message, typica 2026-09-09
commit 0.00 proxy: normalize namespaced commands in Full Access (#18331) Brief domain-specific commit message wit 2026-09-09
commit 0.00 openai: accept plaintext-labeled Codex agent messages (#1832 Terse, technically precise commit messag 2026-09-09
commit 0.00 mlxrunner: compile structured output as xgrammar structural Concise, domain-specific engineering pro 2026-09-03
commit 0.00 mlxrunner: enable speculative decoding under structured outp Specific implementation details and ters 2026-08-29
commit 0.00 gemma4: image and audio input support Specific MLX and Gemma4 implementation d 2026-09-02
commit 0.00 mlxrunner: shared audio decoding and chunking for audio mode Detailed audio constraints and domain-sp 2026-09-02
commit 0.00 mlxrunner: check every mlx-c call in the MLX bindings Terse commit style and precise binding d 2026-09-02
commit 0.00 mlxrunner: capture MLX errors in a single buffer read after Specific error-handling changes and test 2026-09-02
commit 0.00 mlx: fix mtp_test unit test (#18203) Terse conventional commit message with i 2026-09-02
commit 0.00 llama.cpp: version bump b10760 (#18199) Standard concise version-bump commit mes 2026-09-02
commit 0.00 mlx: dedup dependency files (#18159) — 2026-09-01
commit 0.00 lint fix (#18081) — 2026-08-28
commit 0.00 Clean up dead code (#17381) — 2026-08-27
commit 0.00 app: list account cloud models for Claude (#18077) — 2026-08-27
commit 0.00 app: synchronize macOS app handoff (#18056) — 2026-08-27
commit 0.00 mlx: remove the text-only gemma3 model Detailed domain-specific explanation wit 2026-08-25
commit 0.00 mlxrunner: avoid Metal GPU timeouts when loading models from Highly specific Metal, storage, and benc 2026-08-25
commit 0.00 mlxrunner: add structured output support Domain-specific implementation and bench 2026-08-14
commit 0.00 mlx: return exact types and evaluate arrays in the value rea Concise, implementation-specific explana 2026-08-18
commit 0.00 llm: remove the unused Grammar completion-request field Terse, repository-specific cleanup descr 2026-08-13
commit 0.00 llama.cpp: version bump b10630 (#18003) Minimal conventional version-bump messag 2026-08-26
commit 0.00 MLX: version bump (#17955) Minimal conventional version-bump messag 2026-08-26
pr 0.00 mlx: fix windows dependency path resolution The title-only message provides no evide 2026-03-30