26 İşlemeler (4e0d85205a62fc52d44a7fd065f422f1329a6561)

Yazar SHA1 Mesaj Tarih
  orangels 70ced01112 feat: record response debug cases in reports 1 ay önce
  orangels a15a62264a fix: set bench max tokens default to 64k 1 ay önce
  orangels 739524628c feat: report effective benchmark request params 1 ay önce
  orangels 986dc4e088 fix: restore bench max tokens default 1 ay önce
  orangels 1a98836b94 fix: set bench max tokens default to 200k 1 ay önce
  orangels 948e32d64a feat: log request errors for debug 1 ay önce
  orangels e0aca08f36 feat: align gpqa scoring with simple evals 1 ay önce
  orangels 898d02f7ef feat: record upstream error debug payloads 1 ay önce
  orangels 18cc7ff716 feat: add provider thinking parameters 1 ay önce
  orangels ee13739c5e feat: add raw response debug mode 1 ay önce
  orangels 60cc009c83 fix: support bare stream flag and cover SSE parsing 1 ay önce
  orangels f21ab8bd4c feat: add --max-tokens CLI param and correct_samples to bench reports 1 ay önce
  orangels 4988a86654 feat: add progress bars to benchmark and RPM execution 1 ay önce
  orangels 527b1e8fc7 feat: add params field to RPM and benchmark reports 1 ay önce
  orangels fdb437a732 fix: correct streaming bugs and RPM diagnose gaps 1 ay önce
  orangels 20217b2250 feat: add streaming (SSE) support with TTFT metrics 1 ay önce
  orangels 2b9fae7ebd feat: report rpm limiter mode details 1 ay önce
  orangels 0a96e11e98 feat: run rpm limiter modes 1 ay önce
  orangels 65ce102004 fix: resolve only selected provider env refs 1 ay önce
  orangels 31c41cb8ac fix: satisfy final clippy checks 1 ay önce
  orangels 83edc789df fix: make benchmark and rpm execution reliable 1 ay önce
  orangels 5d1163d635 feat: run benchmark and rpm tests 1 ay önce
  orangels 1e0926cfba feat: add openai and anthropic request adapters 1 ay önce
  orangels 2064c69924 fix: tighten benchmark dataset helpers 1 ay önce
  orangels 67a4ea8af4 feat: add benchmark dataset fetch and prompts 1 ay önce
  orangels c0ebdea077 chore: initialize cli project skeleton 1 ay önce