| COMMIT |
1.00 |
cmd/launch: add Copilot CLI integration (#15583) |
|
Commit message contains explicit AI assi |
2026-04-16 |
| COMMIT |
1.00 |
fix: improve error message for unknown input item type in re |
|
Commit message contains explicit AI assi |
2026-04-09 |
| PR |
1.00 |
Optimize ggml_vec_dot_q4_K_q8_K with AVX-512 implementation |
|
Explicit 'code change assisted by Claude |
2026-04-19 |
| PR |
1.00 |
Add killport to Terminal & CLI community integrations |
|
PR body explicitly mentions AI collabora |
2026-04-19 |
| PR |
1.00 |
Reduce memory allocations by 86-99% through buffer reuse opt |
|
Explicit 'Changes are done with Opencode |
2026-04-19 |
| PR |
0.60 |
fix: resolve OpenClaw gateway launch timeout on Windows by e |
|
Polite, structured, and verbose explanat |
2026-04-21 |
| PR |
0.60 |
ADD: OLLAMA_LLM_DEFAULT |
|
Polished, formal tone and generic DevOps |
2024-12-06 |
| PR |
0.40 |
fix(server): implement robust truncation loop for embeddings |
|
Partial template structure, but mostly t |
2026-02-13 |
| PR |
0.30 |
feat: configure auto startup in macos |
|
Somewhat formal but includes specifics a |
2024-10-04 |
| PR |
0.20 |
Install Ollama on OSTree systems |
|
Informal tone, some run-ons, but no AI s |
2024-04-12 |
| PR |
0.20 |
x/tools/mcp: add MCP protocol support for external tool serv |
|
Technical summary and configuration exam |
2026-01-13 |
| PR |
0.20 |
Add Safetensor Conversion for Granite Models |
|
Mostly technical with human review fragm |
2025-01-06 |
| PR |
0.20 |
Fix Ollama silently failing on extra, unsupported openai par |
|
Technical, block code, and natural expla |
2024-09-25 |
| PR |
0.20 |
update langchain-python-rag-document example to use latest l |
|
Mildly formal 'Hello!' but otherwise hum |
2024-05-20 |
| PR |
0.20 |
Refactor app.tsx and index.ts to use APP_DISPLAY_NAME consta |
|
Slightly formal, reads like a good PR, n |
2024-10-04 |
| PR |
0.20 |
[API Client] Feature: Custom HTTP headers for every request |
|
Slightly formal but purpose/use case is |
2023-12-30 |
| PR |
0.15 |
docs: fix tar decompression flags |
|
Casual phrasing and example; lacks AI/fo |
2026-04-19 |
| PR |
0.15 |
openai: honor reasoning_effort in /v1/responses endpoint |
|
Technical explanation with code and conc |
2026-04-18 |
| PR |
0.12 |
app: check user's login shell PATH before prompting for CLI |
|
Technical summary, uses platform specifi |
2026-04-12 |
| COMMIT |
0.10 |
create: avoid gc race with create (#15628) |
|
Technical commit message with informal t |
2026-04-16 |
| COMMIT |
0.10 |
gemma4: render differently based on model size |
|
Uses domain jargon and an informal chang |
2026-04-15 |
| COMMIT |
0.10 |
gemma4: restore e2b-style nothink prompt (#15560) |
|
Technical, concise, uses domain-specific |
2026-04-13 |
| COMMIT |
0.10 |
mac: prevent generate on cross-compiles (#15120) |
|
Concise, informal reasoning with domain- |
2026-04-13 |
| COMMIT |
0.10 |
mlx: add op wrappers for Conv2d, Pad, activations, trig, and |
|
Domain jargon, commit-style writing, not |
2026-04-13 |
| COMMIT |
0.10 |
mlx: mixed-precision quant and capability detection improvem |
|
Technical changelog, uses domain languag |
2026-04-13 |
| COMMIT |
0.10 |
gemma4: fix compiler error on metal (#15550) |
|
Technical and concise, clear bugfix moti |
2026-04-13 |
| COMMIT |
0.10 |
gemma4: update renderer to match new jinja template (#15490) |
|
Informal tone, domain content, brief lis |
2026-04-10 |
| COMMIT |
0.10 |
docs: add hermes agent integration guide (#15488) |
|
Mostly changelog and technical, no AI si |
2026-04-10 |
| COMMIT |
0.10 |
model/parsers: suppress extra gemma4 closing tool tags (#153 |
|
Technical description with informal phra |
2026-04-06 |
| COMMIT |
0.10 |
model/parsers: rework gemma4 tool call handling (#15306) |
|
Concise, technical, uses domain terms, n |
2026-04-03 |
| COMMIT |
0.10 |
server: preserve raw manifest bytes during pull (#15104) |
|
Detailed but technical and focused, like |
2026-03-27 |
| PR |
0.10 |
cmd: populate model capabilities in launchInteractiveModel |
|
Technical and concise with some typos; n |
2026-04-19 |
| PR |
0.10 |
MLX: add repeatable model porting workflow |
|
Informal, some typos, domain-specific no |
2026-04-13 |
| PR |
0.10 |
app/server: fix desktop app startup killing active `ollama l |
|
Technical tone with domain details and c |
2026-04-17 |
| PR |
0.10 |
feat: add option to specify runner name and path in env |
|
Code snippets, direct instructions, and |
2024-12-11 |
| PR |
0.10 |
Ollama docker usage for jetson devices aded to documentation |
|
Concise, informal, with a typo; human-wr |
2024-12-13 |
| PR |
0.10 |
doc: Adding docs on how to compile ollama to run on Intel di |
|
Slightly formal commit body, but no AI s |
2024-10-03 |
| PR |
0.10 |
server: add "Cache-Control: max-age=0" response header |
|
Technical writing with concise human exp |
2024-09-29 |
| PR |
0.10 |
api: add Client.BaseURL method |
|
Technical context and reasoning, not AI |
2024-09-03 |
| PR |
0.10 |
create: keep pending blobs fresh until manifest write |
|
Technical summary, lists, and rationale |
2026-04-17 |
| PR |
0.10 |
Add Spring AI Playground to Chat Interfaces (Desktop) |
|
Mostly list format; free-text is technic |
2026-04-20 |
| PR |
0.10 |
feat: add PARAMETER think support for Modelfile |
|
Slightly formal, but detailed and focuse |
2026-02-06 |
| PR |
0.09 |
install: check for AMD GPU on hybrid systems before exiting |
|
Structured explanation with scenario tab |
2026-04-19 |
| PR |
0.08 |
Create safetensors models through a hybrid local/remote pipe |
|
Domain jargon and brief technical summar |
2026-03-19 |
| PR |
0.08 |
Add LumaKit to Community Integrations |
|
Concise additions, domain-specific, no A |
2026-04-19 |
| PR |
0.08 |
discover: account for reclaimable page cache in cgroup v2 me |
|
Problem/Root cause detail, technical con |
2026-04-19 |
| PR |
0.08 |
Adding a search command |
|
Informal, domain-specific with some typo |
2025-03-29 |
| PR |
0.08 |
Add conversation context usage indicator |
|
Domain-specific, structured and concise; |
2026-03-27 |
| PR |
0.07 |
model/parsers/qwen3coder: prevent tag regex from matching ac |
|
Clear problem explanation with domain te |
2026-04-19 |
| PR |
0.07 |
model/renderers/gemma4: allow reserved JSON Schema keys as p |
|
Technical bugfix, domain terms; free tex |
2026-04-19 |
| PR |
0.06 |
docs(linux): add --zstd flag to tar commands for .tar.zst ar |
|
Brief, direct update with error context; |
2026-04-19 |
| PR |
0.06 |
docs: add community integration to README |
|
Informal and personal tone, typos, typic |
2026-03-24 |
| COMMIT |
0.05 |
bench: add prompt calibration, context size flag, and NumCtx |
|
Technical detail, terse, domain-specific |
2026-04-02 |
| COMMIT |
0.05 |
tokenizer: add byte fallback for SentencePiece BPE encoding |
|
Technical, specific, informal commit mes |
2026-04-02 |
| COMMIT |
0.05 |
Add support for gemma4 (#15214) |
|
Mixture of technical summary and test de |
2026-04-02 |
| COMMIT |
0.05 |
anthropic: fix empty inputs in content blocks (#15105) |
|
Brief, uses domain-appropriate Go termin |
2026-03-27 |
| COMMIT |
0.05 |
launch: skip context length warning for MLX models and show |
|
Concise, omits extraneous detail, classi |
2026-03-27 |
| PR |
0.05 |
mlx: Support NVIDIA TensorRT Model Optimizer import |
|
Short, domain-specific; minimal free-tex |
2026-04-14 |
| PR |
0.05 |
fix: use MemAvailable equivalent in cgroup memory check |
|
Human-style technical explanation with d |
2026-04-20 |
| PR |
0.05 |
ggml: update to 0beb8db3a0 |
|
Structured changelog, strongly human com |
2026-03-15 |
| PR |
0.05 |
Add API key authentication for Ollama server |
|
Concise, list of changes; no AI phrasing |
2025-02-15 |
| PR |
0.05 |
server: fix download stall watchdog not firing when no bytes |
|
Detailed technical breakdown with direct |
2026-04-20 |
| PR |
0.05 |
add MoltenVK support for Intel macs |
|
Casual, personal, minor grammar flaws; c |
2026-03-22 |
| PR |
0.05 |
runner/ollamarunner: handle sampling errors gracefully |
|
Technical, includes typo, brief phrasing |
2026-03-17 |
| PR |
0.01 |
ml/backend/ggml: add cascadelake ggml build config |
|
Human, terse, references specific issues |
2026-02-26 |
| COMMIT |
0.00 |
launch: replace kimi-k2.5 with k2.6 as top recommended model |
|
Short, technical commit summary; human s |
2026-04-21 |
| COMMIT |
0.00 |
mlxrunner: tokenize prompts in request handler goroutines |
|
Detailed technical writing with domain j |
2026-04-03 |
| COMMIT |
0.00 |
mlx: improve thread safety of array management |
|
Technical, detailed and with inline rema |
2026-04-03 |
| COMMIT |
0.00 |
app/ui: fix model picker showing stale model after switching |
|
Includes bug details and terse explanati |
2026-04-21 |
| COMMIT |
0.00 |
cmd: populate model capabilities in launchInteractiveModel ( |
|
References prior PR and bug, highly spec |
2026-04-21 |
| COMMIT |
0.00 |
mlxrunner: fuse top-P and top-K into a single sort pass |
|
Technical improvement with domain vocabu |
2026-04-16 |
| COMMIT |
0.00 |
mlxrunner: use MaxAxis in the min-P sampler |
|
Very terse, technical, domain-specific; |
2026-04-16 |
| COMMIT |
0.00 |
mlxrunner: add logprobs support |
|
Concise, domain-specific; matches human |
2026-04-16 |
| COMMIT |
0.00 |
server: apply format when think=false for gemma4 (#15678) |
|
One-line ticket close, matches accepted |
2026-04-21 |
| COMMIT |
0.00 |
launch: add kimi cli integration with installer flow (#15723 |
|
Concise commit message; no AI hallmarks. |
2026-04-20 |
| COMMIT |
0.00 |
mlx: fuse sigmoid router head in glm4_moe_lite |
|
Technical, domain-specific; human-writte |
2026-04-15 |
| COMMIT |
0.00 |
mlx: apply repeat penalties in sampler (#15631) |
|
Very terse commit message, classic human |
2026-04-18 |
| COMMIT |
0.00 |
docs: update hermes (#15655) |
|
Standard doc update commit; no AI signal |
2026-04-17 |
| COMMIT |
0.00 |
cmd/launch: show WSL guidance on Windows instead of handing |
|
Terse, domain-specific commit; no AI mar |
2026-04-17 |
| COMMIT |
0.00 |
launch: skip unchanged managed-single rewrite (#15633) |
|
Concise, technical language with no AI c |
2026-04-16 |
| COMMIT |
0.00 |
mlx: additional gemma4 cache fixes (#15607) |
|
Brevity, domain-specific, informal; no A |
2026-04-16 |
| COMMIT |
0.00 |
mlx: fix gemma4 cache to use logical view (#15617) |
|
Short, terse, and domain-specific; no AI |
2026-04-16 |
| COMMIT |
0.00 |
mlx: fix imagegen lookup (#15588) |
|
Domain-specific, includes references/typ |
2026-04-16 |
| COMMIT |
0.00 |
make empty block conditional |
|
Extremely brief and direct; no sign of A |
2026-04-15 |
| COMMIT |
0.00 |
Keep Gemma4 router projection in source precision (#15613) |
|
Short technical title, no AI hallmarks o |
2026-04-15 |
| COMMIT |
0.00 |
launch: always list cloud recommendations first (#15593) |
|
Concise, informal, clearly human-written |
2026-04-15 |
| COMMIT |
0.00 |
launch: add hermes (#15569) |
|
Brief technical commit, no signs of AI p |
2026-04-15 |
| COMMIT |
0.00 |
mlx: fix RotatingKVCache.concat() dropping context on mid-ro |
|
Detailed technical commit, domain jargon |
2026-04-15 |
| COMMIT |
0.00 |
mlx: Improve gemma4 performance with fused operations (#1558 |
|
Changelog format, domain-specific, no AI |
2026-04-15 |
| COMMIT |
0.00 |
models: fuse MLP activation functions via mlx_compile |
|
Technical description, domain-specific, |
2026-04-13 |
| COMMIT |
0.00 |
mlx: add compiled closure support |
|
Technical and concise, structure is typi |
2026-04-13 |
| COMMIT |
0.00 |
launch: OpenCode inline config (#15586) |
|
Concise commit message with standard hum |
2026-04-14 |
| COMMIT |
0.00 |
launch/openclaw: fix --yes flag behaviour to skip channels c |
|
Standard technical fix description, ters |
2026-04-14 |
| COMMIT |
0.00 |
Revert "launch/opencode: use inline config (#15462)" (#15568 |
|
Typical revert message from GitHub, no A |
2026-04-14 |
| COMMIT |
0.00 |
launch: skip unchanged integration rewrite configration (#15 |
|
Brief summary with a typo, likely human- |
2026-04-14 |
| COMMIT |
0.00 |
Gemma4 on MLX (#15244) |
|
Technical changelog with informal and do |
2026-04-13 |
| COMMIT |
0.00 |
launch/opencode: use inline config (#15462) |
|
Pure changelog, no free-text aside from |
2026-04-13 |
| COMMIT |
0.00 |
Revert "gemma4: fix nothink case renderer (#15553)" (#15556) |
|
Standard revert message, no free-text ev |
2026-04-13 |
| COMMIT |
0.00 |
cgo: suppress deprecated warning to quiet down go build (#15 |
|
Changelog style, no evidence of AI-gener |
2026-04-13 |
| COMMIT |
0.00 |
Revert "gemma4: add nothink renderer tests (#15554)" (#15555 |
|
Template revert, no free-text for AI ana |
2026-04-13 |
| COMMIT |
0.00 |
gemma4: add nothink renderer tests (#15554) |
|
Brief and project-specific explanation, |
2026-04-13 |
| COMMIT |
0.00 |
gemma4: fix nothink case renderer (#15553) |
|
Terse, domain-specific commit message wi |
2026-04-13 |
| COMMIT |
0.00 |
Update to ROCm 7.2.1 (#15483) |
|
Commit uses a human co-author GitHub tra |
2026-04-12 |
| COMMIT |
0.00 |
gemma4: be less strict about whitespace before bare keys (#1 |
|
Terse, informal style; clear domain cont |
2026-04-11 |
| COMMIT |
0.00 |
model/parsers: fix missing parallel tool call indices (#1546 |
|
Direct changelog style, terse; clearly h |
2026-04-10 |
| COMMIT |
0.00 |
app/ui: re-validate image attachments when selected model ch |
|
No free-text, only title; human or templ |
2026-04-10 |
| COMMIT |
0.00 |
launch: update openclaw channel message (#15463) |
|
Very terse commit message, no stylistic |
2026-04-09 |
| COMMIT |
0.00 |
launch: add re-run hint to dependency error message (#15439) |
|
Terse, domain-specific and informal phra |
2026-04-09 |
| COMMIT |
0.00 |
modelfiles: fix /save command and add shortname for safetens |
|
Contains domain detail, numbered issues; |
2026-04-09 |
| COMMIT |
0.00 |
mlx: remove stale x86 libmlx library (#15443) |
|
Short, informal; clear reference to prio |
2026-04-09 |
| COMMIT |
0.00 |
app: restore launch default and refine launch sidebar open f |
|
Brief, technical, and lacks AI-typical s |
2026-04-08 |
| COMMIT |
0.00 |
pull/push: refine safetensors (#14946) |
|
Domain-specific details, informal change |
2026-04-08 |
| COMMIT |
0.00 |
launch/opencode: detect `curl` installed opencode at `~/.ope |
|
No free-text, terse and technical commit |
2026-04-08 |
| COMMIT |
0.00 |
launch: add openclaw channels setup (#15407) |
|
No free-text, direct technical summary. |
2026-04-08 |
| COMMIT |
0.00 |
launch: update ctx length for glm-5.1 and gemma4 (#15411) |
|
Brief, technical, with human-style conte |
2026-04-08 |
| COMMIT |
0.00 |
cmd: improve multi-select sorting and selection status (#152 |
|
Short, informal summary, no AI-like phra |
2026-04-08 |
| COMMIT |
0.00 |
responses: add support for fn call output arrays (#15406) |
|
Technical with error logs, informal and |
2026-04-07 |
| COMMIT |
0.00 |
app/ui: fix lint errors for unused vars, prefer-const, and e |
|
Changelog style, terse and technical. |
2026-04-07 |
| COMMIT |
0.00 |
gemma4: Disable FA on older GPUs where it doesn't work (#154 |
|
Brief, technical commit with domain cont |
2026-04-07 |
| COMMIT |
0.00 |
mlx: use default http client (#15405) |
|
No free-text commentary, only standard s |
2026-04-07 |
| COMMIT |
0.00 |
gemma4: add missing file (#15394) |
|
Succinct explanation, normal for human c |
2026-04-07 |
| COMMIT |
0.00 |
create: Clean up experimental paths, fix create from existi |
|
Structured changelog with terse explanat |
2026-04-07 |
| COMMIT |
0.00 |
gemma4: enable flash attention (#15378) |
|
Technical terminology and direct phrasin |
2026-04-07 |
| COMMIT |
0.00 |
mlx: Improve M5 performance with NAX (#15345) |
|
Expansive technical commit, uses domain- |
2026-04-07 |
| COMMIT |
0.00 |
model/parsers: add gemma4 tool call repair (#15374) |
|
Technical explanation with domain detail |
2026-04-07 |
| COMMIT |
0.00 |
launch/openclaw: patch approvedScopes baseline for TUI pairi |
|
No free-text, just a standard, short com |
2026-04-07 |
| COMMIT |
0.00 |
app: update featured models (#15373) |
|
Concise task description, typical of hum |
2026-04-06 |
| COMMIT |
0.00 |
app: default app home view to new chat instead of launch (#1 |
|
Terse commit message, informal and speci |
2026-04-04 |
| COMMIT |
0.00 |
Revert "enable flash attention for gemma4 (#15296)" (#15311) |
|
Standard git revert message, no AI signa |
2026-04-04 |
| COMMIT |
0.00 |
ggml: fix ROCm build for cublasGemmBatchedEx reserve wrapper |
|
Direct, technical fix description, typic |
2026-04-03 |
| COMMIT |
0.00 |
enable flash attention for gemma4 (#15296) |
|
Terse PR summary, lacks AI-like formalit |
2026-04-03 |
| COMMIT |
0.00 |
ggml: skip cublasGemmBatchedEx during graph reservation |
|
Technical explanation, abbreviations, in |
2026-04-03 |
| COMMIT |
0.00 |
model/parsers: fix gemma4 arg parsing when quoted strings co |
|
Multiple commits, technical content, cle |
2026-04-03 |
| COMMIT |
0.00 |
app: use the same client for inference and other requests (# |
|
Concise, plain commit message; no AI tex |
2026-04-02 |
| COMMIT |
0.00 |
app: add launch page for a simple way to launch integrations |
|
Very brief, no stylistic evidence of AI |
2026-04-02 |
| COMMIT |
0.00 |
launch: replace deprecated OPENAI_BASE_URL with config.toml |
|
Brief, domain-specific commit message; n |
2026-04-01 |
| COMMIT |
0.00 |
tokenizer: add SentencePiece-style BPE support (#15162) |
|
Contains review comments and technical j |
2026-04-01 |
| COMMIT |
0.00 |
mlx: respect tokenizer add_bos_token setting in pipeline (#1 |
|
Thorough technical changelog, specific j |
2026-03-31 |
| COMMIT |
0.00 |
docs: update pi docs (#15152) |
|
Typical doc update summary; too brief fo |
2026-03-31 |
| COMMIT |
0.00 |
cmd: set OpenCode default model in config (#15127) |
|
Concise commit message, entirely human s |
2026-03-29 |
| COMMIT |
0.00 |
model: add qwen3-next compatibility for legacy ssm_in projec |
|
Brief, domain-specific commit; no AI sig |
2026-03-29 |
| COMMIT |
0.00 |
revert context length warnings change (#15121) |
|
Terse, typical revert commit phrasing. |
2026-03-28 |
| COMMIT |
0.00 |
launch: improve multi-select for already added models (#1511 |
|
Standard commit summary, no AI signals. |
2026-03-28 |
| COMMIT |
0.00 |
launch: auto-install pi and manage web-search lifecycle (#15 |
|
Succinct and technical, no AI phrasing d |
2026-03-28 |
| COMMIT |
0.00 |
mlx: fix vision capability + min version (#15106) |
|
Typical terse commit style, no AI indica |
2026-03-28 |
| COMMIT |
0.00 |
anthropic: fix KV cache reuse degraded by tool call argument |
|
Technical explanation, domain jargon, no |
2026-03-09 |
| PR |
0.00 |
server: add v2 manifest path |
|
Structured, technical, with domain detai |
2026-04-21 |
| PR |
0.00 |
launch: replace kimi-k2.5 with kimi-k2.6 as top recommended |
|
Single sentence; entirely human and conc |
2026-04-21 |
| PR |
0.00 |
mlxrunner: batch the sampler across multiple sequences |
|
Dense technical detail and jargon; no AI |
2026-04-21 |
| PR |
0.00 |
kvcache: add TurboQuant compressed KV cache (tq2/tq3/tq2k/tq |
|
Technical summary with table and specifi |
2026-04-11 |
| PR |
0.00 |
mlxrunner: tokenize prompts in request handler goroutines |
|
Technical explanation, clear intent, no |
2026-04-01 |
| PR |
0.00 |
app/ui: fix model picker showing stale model after switching |
|
Technical, with specific bug explanation |
2026-04-03 |
| PR |
0.00 |
docs: Structured outputs and ollama cloud |
|
No content; certainly human/empty. |
2026-04-21 |
| PR |
0.00 |
Add model eval metrics to /metrics |
|
Largely a direct changelog with code blo |
2025-06-22 |
| PR |
0.00 |
add conversion code for cohere2 arch |
|
Terse, technical, minimal free-text; no |
2025-01-16 |
| PR |
0.00 |
fix: only add to history if different |
|
Brief, informal, domain-specific and min |
2024-12-21 |
| PR |
0.00 |
server: print environment variables in sorted, one-line |
|
Direct, technical statement with no AI s |
2024-12-06 |
| PR |
0.00 |
Document that `--format` now supports passing JSON Schemas |
|
Casual clarifying question; natural huma |
2024-12-06 |
| PR |
0.00 |
api: change /delete endpoint to use POST method |
|
Concise, typical PR description; no AI s |
2024-12-09 |
| PR |
0.00 |
Delete redundancy code when never happends. |
|
Terse, informal language with grammar er |
2024-12-09 |
| PR |
0.00 |
cmd: Add --base2 option to ps to show model sizes in KiB/MiB |
|
Contains domain-specific output and info |
2024-12-11 |
| PR |
0.00 |
Fix message truncation logic and ensure at least one system |
|
Standard templated changelog, straightfo |
2024-12-06 |
| PR |
0.00 |
Clarify num_ctx is measured in tokens |
|
Direct, concise clarification; lacks AI |
2024-12-03 |
| PR |
0.00 |
Fix /api/show map options bug. |
|
Minimalistic, repetitive bug fix message |
2024-10-25 |
| PR |
0.00 |
cmd: make func addToZipFile |
|
Short, technical, and focused; lacks AI |
2024-11-30 |
| PR |
0.00 |
Apply some gosec rules |
|
Brief, structured change log; human engi |
2024-11-27 |
| PR |
0.00 |
Export ctx, gpu, parallel parameters via /api/ps |
|
Brief technical summary, no AI-like phra |
2024-11-26 |
| PR |
0.00 |
added formatting to ollama ls |
|
Informal tone with technical detail and |
2024-10-20 |
| PR |
0.00 |
server: fix custom template capabiliy checking for generatio |
|
Technical explanation, incomplete senten |
2024-09-29 |
| PR |
0.00 |
Expose ROCm P2P workaround as runtime config |
|
Verbose but human-like technical analysi |
2024-10-26 |
| PR |
0.00 |
Fix openapi base writer header code. |
|
Terse, repetitive commit message, typica |
2024-10-14 |
| PR |
0.00 |
Update RAM requirements for model sizes in README.md |
|
Colloquial language and minor grammar er |
2024-10-13 |
| PR |
0.00 |
Fixed no redirect URL scenario when downloading blobs |
|
Very brief bug fix message, normal for h |
2024-09-11 |
| PR |
0.00 |
Add truncation guard |
|
Very concise, informal and specific huma |
2024-05-24 |
| PR |
0.00 |
Allow XDG user directories |
|
Text includes informal tone and user adm |
2024-05-12 |
| PR |
0.00 |
docs: Add dnf install "hipblas rocm-*" to Linux Installation |
|
Short, context reference only, no AI sig |
2024-05-19 |
| PR |
0.00 |
feat: ollama run --parameters/-p |
|
Informal 'Hey', specific usage example, |
2024-03-14 |
| PR |
0.00 |
server: apply format when think=false with thinking-capable |
|
Structured, technical; concise test plan |
2026-04-18 |
| PR |
0.00 |
convert: support fp8 safetensors import |
|
Complex domain details, lists; strongly |
2026-04-20 |
| PR |
0.00 |
MLX Sampler Improvements |
|
Technical bullets, reviewer input; stron |
2026-04-17 |
| PR |
0.00 |
launch: add kimi cli integration with installer flow |
|
Brief, plain description; no AI traits. |
2026-04-20 |
| PR |
0.00 |
mlx: fuse sigmoid router head in glm4_moe_lite |
|
Highly technical, compact; matches domai |
2026-04-18 |
| PR |
0.00 |
launch: fetch recommended models from server endpoint |
|
Well-structured, technical; matches stan |
2026-04-14 |
| PR |
0.00 |
runner: Remove CGO engines, use llama-server exclusively for |
|
WIP notice, casual tone, specific contex |
2026-03-28 |
| PR |
0.00 |
fix(docs): replace incorrect tr command in Linux uninstall i |
|
Technical, specific, and concise changel |
2026-04-19 |
| PR |
0.00 |
model/models/nomicbert: fix tokenizer for t5 models (crash + |
|
Highly detailed, domain-specific and inf |
2026-04-18 |
| PR |
0.00 |
add token calculation support with UI display (#15639) |
|
Contains domain jargon and example code; |
2026-04-19 |
| PR |
0.00 |
fix: prevent head-of-line blocking in scheduler during model |
|
Casual, technical explanation with jargo |
2026-03-29 |
| PR |
0.00 |
MLX Closures |
|
No content beyond title; insufficient da |
2026-04-14 |
| PR |
0.00 |
server: preserve thinking in /api/generate and populate para |
|
Bug descriptions, informal and technical |
2026-04-18 |
| PR |
0.00 |
llm: report runner status when Wait returns nil |
|
Direct, technical explanation with domai |
2026-04-18 |
| PR |
0.00 |
docs: canonicalize target-server-local documentation |
|
Dense, technical summary with validation |
2026-04-18 |
| PR |
0.00 |
fix(metal): resolve static_assert crash on Apple M5 |
|
Direct reference to hardware, upstream P |
2026-03-04 |
| PR |
0.00 |
mlx: apply repeat penalties in sampler |
|
Very short, technical commit; no AI phra |
2026-04-16 |
| PR |
0.00 |
fix(windows): prevent system sleep during model downloads an |
|
Technical bullet points, direct and cont |
2026-04-18 |
| PR |
0.00 |
fix: strip surrounding quotes from drag-dropped file paths i |
|
Problem-focused, includes realistic term |
2026-04-18 |
| PR |
0.00 |
docs: update hermes |
|
Very short free text, informal and human |
2026-04-17 |
| PR |
0.00 |
cmd: guard TUI launch against non-interactive terminals |
|
Technical, concise commit describing a b |
2026-04-17 |
| PR |
0.00 |
cmd: improve CLI error messages for missing model name arg |
|
Mostly a template with no free-text; no |
2026-04-17 |
| PR |
0.00 |
cmd: improve CLI error messages for missing model name arg |
|
Concise, domain-specific, command list; |
2026-04-17 |
