Commit Graph

11 Commits

Author SHA1 Message Date
Colton Padden 96c01c3d26 docs(benchmarks): publish Astra, Gemini 3.8, Minimax M3, and Kimit k2 model results (#3380)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-09-17 11:01:40 -04:00
Colton Padden 577d78b5d9 Benchmark Claude Fable 5.1 and GLM 5.3 (#2884)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-09-02 11:17:23 -04:00
Colton Padden 7cf2603823 docs(benchmarks): publish codex harness results (#2788)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-08-31 17:01:28 -04:00
Colton Padden 32d2984676 refactor(benchmarks): use native canary runners and upgrade to agent-eval v2 (#2785)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-08-31 17:01:28 -04:00
Colton Padden 83d40081dd docs(benchmarks) anthropic to use claude harness introduce opus 5 (#2731)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-08-31 13:11:09 -04:00
Colton Padden 451be3fb2d docs(benchmarks): run benchmarks and retain superseded models (#2724)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-08-31 13:11:09 -04:00
Colton Padden 862b39a751 docs: publish expanded authoring benchmark matrix (#2198)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-08-19 21:32:57 -04:00
OwenKephart 5678b9ee07 perf: end benchmark turns when the agent is done (#2249)
Signed-off-by: Owen Kephart <owen.kephart@vercel.com>
Signed-off-by: owenkephart <owen.kephart@vercel.com>
2026-08-18 13:36:19 -07:00
OwenKephart 085aaee999 perf: profile authoring benchmark setup (#2219)
Signed-off-by: owenkephart <owen.kephart@vercel.com>
2026-08-18 13:36:18 -07:00
Colton Padden 860ad6538b feat: add authoring benchmark model matrix (#2111)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-08-17 20:37:49 -04:00
Colton Padden 4f78a2d336 feat: publish authoring benchmark results (#2057)
Signed-off-by: Colton Padden <colton.padden@vercel.com>
2026-08-17 20:22:12 -04:00