Default branch

main
Some checks are pending
CI / Detect changes (push) Waiting to run
CI / Test Suite (push) Blocked by required conditions
CI / Test Suite-1 (push) Blocked by required conditions
CI / Test Suite-2 (push) Blocked by required conditions
CI / Rustfmt (push) Blocked by required conditions
CI / Clippy (push) Blocked by required conditions
CI / Cargo Check (push) Blocked by required conditions
CI / Python Checks (push) Blocked by required conditions
Community Benchmarks / Validate submissions (push) Waiting to run
Release Please / release-please (push) Waiting to run

eb8db38351 · feat(cli): show backend-incompatible models on request (#1143) · Updated 2026-10-15 16:00:37 +00:00

Branches

81ea00385f · chore: update Cargo.lock · Updated 2026-10-15 16:01:38 +00:00

0
2

9c76f32a49 · chore(models): Weekly model data refresh · Updated 2026-10-12 03:09:51 +00:00

21
1

a556dc2aa0 · chore(data): Weekly benchmark cache refresh · Updated 2026-10-12 02:24:23 +00:00

21
1

45b83975a2 · ci(release): say that a backfill re-run submits new signing requests · Updated 2026-10-09 14:46:30 +00:00

24
2

90d7de3092 · test: pin the kernel fallback penalty on one GPU and cover the new JSON fields · Updated 2026-10-09 14:46:11 +00:00

24
2

19d42f1c9d · fix(models): keep decimal sizes intact before a file extension · Updated 2026-10-02 20:13:00 +00:00

51
2

c406a7d372 · fix(providers): map gemma4:12b and other Sept 2026 catalog additions to Ollama tags · Updated 2026-09-19 08:18:53 +00:00

67
1

4d917afcf4 · fix(update): read the context window from config.json · Updated 2026-09-19 07:28:30 +00:00

75
1

90cb737bcd · fix(bench): stop wall timer after body read in OpenAI and Ollama paths · Updated 2026-09-14 06:45:12 +00:00

88
1

b8d8f58790 · fix(tui): update model viewport during event handling · Updated 2026-09-10 05:59:23 +00:00

98
2

12c214fa49 · fix(docker): build each platform natively instead of cross-compiling under QEMU · Updated 2026-09-03 13:17:24 +00:00

108
1

0919219c72 · fix(hardware): recover BIOS UMA carveout on Linux APUs and drop legacy Intel iGPUs (#964) · Updated 2026-09-03 11:14:18 +00:00

110
1

bb3c902500 · fix(hardware): keep Linux BIOS UMA carve-out VRAM on Ryzen AI MAX · Updated 2026-09-03 10:39:03 +00:00

110
1

8a689fe9e7 · fix: validate recurrent layout heuristics · Updated 2026-08-25 12:05:45 +00:00    vrr

159
2

976feb4600 · feat(models): add the Qwen3.8 family and fix Qwen generation parsing · Updated 2026-08-17 06:21:35 +00:00    vrr

192
1

d687a39771 · fix(providers): match every K-quant and I-quant GGUF variant · Updated 2026-08-13 11:35:14 +00:00    vrr

193
1

b4c44a587b · fix(share): explain a refused write to the user's own fork · Updated 2026-08-11 16:30:46 +00:00    vrr

200
1

83d6f685ff · fix(python): read the package readme through the metadata hook · Updated 2026-08-11 12:10:29 +00:00    vrr

201
3

701cff1d01 · fix(cli): don't auto-spawn dashboard for read-only subcommands · Updated 2026-08-07 06:24:12 +00:00    vrr

221
1

931f7b3422 · Merge remote-tracking branch 'origin/main' into steward/810-windows-uma-vram · Updated 2026-08-05 06:04:53 +00:00    vrr

228
2