ffb25b0
(HEAD -> main)
Merge pull request 'codex/lq-token-test-init' (#2) from codex/lq-token-test-init into main by
2026-05-10 13:34:07 +0800
c346233
(refs/pull/2/head, codex/lq-token-test-init)
feat: record stream chunk timings in reports by
2026-05-10 13:25:46 +0800
1f5bdb2
feat: group reports by local date by
2026-05-10 12:00:05 +0800
506434d
fix: use local time for raw debug logs by
2026-05-10 11:55:28 +0800
4e0d852
feat: group raw debug logs by date by
2026-05-10 11:51:08 +0800
70ced01
feat: record response debug cases in reports by
2026-05-10 11:43:09 +0800
a15a622
fix: set bench max tokens default to 64k by
2026-05-09 21:26:42 +0800
1243558
fix: include partial stream in interrupted debug logs by
2026-05-09 20:14:03 +0800
7395246
feat: report effective benchmark request params by
2026-05-09 20:05:33 +0800
91e863b
Merge pull request 'codex/lq-token-test-init' (#1) from codex/lq-token-test-init into main by
2026-05-09 19:38:02 +0800
986dc4e
(refs/pull/1/head)
fix: restore bench max tokens default by
2026-05-09 18:55:32 +0800
b5d83f9
fix: log empty successful responses as request errors by
2026-05-09 18:42:37 +0800
1a98836
fix: set bench max tokens default to 200k by
2026-05-09 18:30:40 +0800
948e32d
feat: log request errors for debug by
2026-05-09 18:12:30 +0800
f17c0af
feat: tighten aime final answer parsing by
2026-05-09 17:51:54 +0800
e0aca08
feat: align gpqa scoring with simple evals by
2026-05-09 17:37:09 +0800
898d02f
feat: record upstream error debug payloads by
2026-05-09 16:21:25 +0800
5706ac8
feat: add google gemini protocol by
2026-05-09 15:58:04 +0800
6959de5
docs: add CLAUDE.md and update config.example.yaml by
2026-05-07 04:57:36 +0800
18cc7ff
feat: add provider thinking parameters by
2026-05-07 04:08:00 +0800
ee13739
feat: add raw response debug mode by
2026-05-07 03:12:37 +0800
60cc009
fix: support bare stream flag and cover SSE parsing by
2026-05-07 02:47:58 +0800
f21ab8b
feat: add --max-tokens CLI param and correct_samples to bench reports by
2026-05-07 02:17:36 +0800
4988a86
feat: add progress bars to benchmark and RPM execution by
2026-05-07 00:19:25 +0800
527b1e8
feat: add params field to RPM and benchmark reports by
2026-05-06 23:55:16 +0800
e269f8f
docs: add end-to-end testing guide for RPM and benchmarks by
2026-05-06 23:32:06 +0800
fdb437a
fix: correct streaming bugs and RPM diagnose gaps by
2026-05-06 23:15:33 +0800
20217b2
feat: add streaming (SSE) support with TTFT metrics by
2026-05-06 22:51:54 +0800
f5b9239
docs: add Chinese usage guide by
2026-05-06 22:14:53 +0800
84943db
docs: document rpm limiter modes by
2026-05-06 21:27:36 +0800
2b9fae7
feat: report rpm limiter mode details by
2026-05-06 21:25:54 +0800
0a96e11
feat: run rpm limiter modes by
2026-05-06 21:22:19 +0800
7c0d1e8
feat: add rpm mode schedules by
2026-05-06 21:18:36 +0800
f6b09d0
docs: plan rpm limiter modes by
2026-05-06 21:16:07 +0800
65ce102
fix: resolve only selected provider env refs by
2026-05-06 20:27:00 +0800
4b433c2
fix: redact config tokens in debug output by
2026-05-06 20:13:02 +0800
31c41cb
fix: satisfy final clippy checks by
2026-05-06 20:07:27 +0800
4e99ab6
docs: add usage guide by
2026-05-06 20:00:45 +0800
83edc78
fix: make benchmark and rpm execution reliable by
2026-05-06 18:10:15 +0800
5d1163d
feat: run benchmark and rpm tests by
2026-05-06 17:59:16 +0800
4a0e187
docs: specify benchmark report output by
2026-05-06 17:33:12 +0800
d0fed7b
fix: harden protocol adapter errors by
2026-05-06 17:22:58 +0800
1e0926c
feat: add openai and anthropic request adapters by
2026-05-06 17:07:43 +0800
2064c69
fix: tighten benchmark dataset helpers by
2026-05-06 16:59:24 +0800
67a4ea8
feat: add benchmark dataset fetch and prompts by
2026-05-06 16:49:45 +0800
690b815
test: avoid global env mutation in config tests by
2026-05-06 16:40:37 +0800
b071c99
feat: load yaml relay config by
2026-05-06 16:33:19 +0800
c0ebdea
chore: initialize cli project skeleton by
2026-05-06 16:18:53 +0800
ea0b255
docs: add implementation plan by
2026-05-06 16:10:30 +0800
61446dc
docs: define benchmark dataset download flow by
2026-05-06 16:04:12 +0800
6e804b9
docs: use official gsm8k benchmark by
2026-05-06 15:54:13 +0800
8840977
docs: add design spec by
2026-05-06 15:50:41 +0800