12月1月2月3月4月5月6月7月8月9月10月11月
周一周三周五
一年内 260365 次贡献
更少更多
- frostbyte_neo 推送了仓库 github-featured/alirezarezvani--claude-skills 的 dev 分支比较 85 提交 »31a23f0191 Merge pull request #961 from alirezarezvani/claude/agent-launcher-plugin-2pd4qc3e29c960fa merge dev into agent-launcher branch: counters trued to 377/695/817/94, D1 sidecar move, models pinned to claude-opus-5, references topped up6cba04a0a5 chore: sync codex skills symlinks [automated]4d7df8adeb Merge pull request #948 from alirezarezvani/claude/humanizer-skill-audit-plugin-hocj85fbc3cdc3f7 merge dev into human-gate branch: resolve counter surfaces, move attribution to authoring-notes.json sidecar (issue #954 policy), true up counters to 371/675/812/93
- frostbyte_neo 推送了仓库 github-featured/alirezarezvani--claude-skills 的 main 分支比较 19 提交 »98180dafc4 Merge pull request #970 from alirezarezvani/dev6972e654ca chore: sync codex skills symlinks [automated]a0133f5193 Merge pull request #947 from alirezarezvani/claude/memory-engineering-skills-uwaifx2c384a9c47 fix(memory-engineering): close an F1 bypass and make --print-sample-spec reachable38cf9be64d docs(readme): true up the reference-docs prose counter 746 -> 750
- frostbyte_neo 推送了仓库 github-featured/bjarneo--cliamp 的 main 分支比较 3 提交 »2ac3f9ef32 fix(ci): restore security and Windows checksa7b9eda1d6 fix(release): scope Go dependency to HEAD1f2e5ac0b5 fix(player): prevent radio stutter and channel skips
- frostbyte_neo 推送了仓库 github-featured/lightseekorg--tokenspeed 的 perf/k3-kda-payload-copy 分支比较 13 提交 »35594c74f2 Merge branch 'main' into perf/k3-kda-payload-copy7a4fb3b88b perf(k3): fuse the speculative round's KDA metadata into single launchesa2daa17d1a feat(kimi-k3): serve DSpark drafts (fc_norm + AttnRes tap) (#1031)2706143a86 Upgrade TokenSpeed Scheduler to 0.1.9 (#1208)455c17d2c6 Bump TokenSpeed Scheduler to 0.1.9 (#1207)
- frostbyte_neo 推送了仓库 github-featured/lightseekorg--tokenspeed 的 ts2/k3-aime26-64k 分支8cfe3d7984 ci(k3): give the Slurm aime26 gates the budget their context window allows
- frostbyte_neo 推送了仓库 github-featured/lightseekorg--tokenspeed 的 ts2/kda-triton-tuning 分支比较 9 提交 »7ff10897c8 Merge branch 'main' into ts2/kda-triton-tuningaae140cf59 perf(kda): unify NVIDIA on the V-major state and split verify producersa2daa17d1a feat(kimi-k3): serve DSpark drafts (fc_norm + AttnRes tap) (#1031)2706143a86 Upgrade TokenSpeed Scheduler to 0.1.9 (#1208)455c17d2c6 Bump TokenSpeed Scheduler to 0.1.9 (#1207)
- frostbyte_neo 推送了仓库 github-featured/lightseekorg--tokenspeed 的 main 分支比较 5 提交 »a2daa17d1a feat(kimi-k3): serve DSpark drafts (fc_norm + AttnRes tap) (#1031)2706143a86 Upgrade TokenSpeed Scheduler to 0.1.9 (#1208)455c17d2c6 Bump TokenSpeed Scheduler to 0.1.9 (#1207)0683caf969 fix(cli): disable smg load monitoring by default (#1205)9f1c9cf016 fix(runtime): make load queries cancellation-safe (#1206)
- frostbyte_neo 推送了仓库 github-featured/lightseekorg--tokenspeed 的 borontion/amd-dsv4-1 分支比较 16 提交 »ac2e952bcb ci: keep DeepSeek V4 platform evals separateb73e0fdb15 ci: consolidate DeepSeek V4 evals4a87cf0985 Cleanup013a4b5eb5 Merge remote-tracking branch 'origin/main' into borontion/amd-dsv4-1a2daa17d1a feat(kimi-k3): serve DSpark drafts (fc_norm + AttnRes tap) (#1031)
- frostbyte_neo 推送了仓库 github-featured/clacky-ai--openclacky 的 main 分支1ce45a310b feat: advisor for next job
- frostbyte_neo 推送了仓库 github-featured/anomalyco--models.dev 的 dev 分支比较 12 提交 »9ae8518bce chore(sync): update NanoGPT model catalog (#5303)e7b9519135 chore(sync): update OpenRouter model catalog (#5302)d4b6ea2913 chore(sync): update OpenRouter model catalog (#5300)ea59dc866f chore(sync): update Kilo model catalog (#5299)9af6eb2658 chore(sync): update Kilo model catalog (#5297)
- frostbyte_neo 推送了仓库 github-featured/piebald-ai--claude-code-system… 的 main 分支比较 2 提交 »cf845eeaaa Update changelog for v2.1.2410260612c8d v2.1.241 (+182 tokens)
- frostbyte_neo 推送了仓库 github-featured/roboflow--rf-detr 的 io/draft 分支比较 2 提交 »8c1d81f3f7 fix(data): preserve JPEG draft geometry753632ce5b Merge branch 'develop' into io/draft
- frostbyte_neo 推送了仓库 github-featured/roboflow--rf-detr 的 gh-pages 分支比较 4 提交 »4ca5b9cb65 Deployed 74bfc449 to develop with MkDocs 1.6.1 and mike 2.2.08c21fa644e Deployed 73867d3c to develop with MkDocs 1.6.1 and mike 2.2.04805720793 Deployed 86773509 to develop with MkDocs 1.6.1 and mike 2.2.07c5fe262f7 Deployed e5be6c4f to develop with MkDocs 1.6.1 and mike 2.2.0
- frostbyte_neo 推送了仓库 github-featured/roboflow--rf-detr 的 perf/coco 分支a42a76a480 docs(cookbooks): tune COCO2017 for high-end GPUs
- frostbyte_neo 推送了仓库 github-featured/roboflow--rf-detr 的 develop 分支比较 4 提交 »74bfc449b4 perf(inference): transfer CUDA source images as bytes (#1388)73867d3c57 perf(inference): skip known-valid image range scans (#1387)86773509e4 fix(datasets): handle Albumentations flip aliases (#1386)e5be6c4f29 perf(transformer): skip single-level deformable-attention packing (#1385)
