orangels
|
739524628c
|
feat: report effective benchmark request params
|
1 tydzień temu |
orangels
|
948e32d64a
|
feat: log request errors for debug
|
1 tydzień temu |
orangels
|
f17c0af5d7
|
feat: tighten aime final answer parsing
|
1 tydzień temu |
orangels
|
e0aca08f36
|
feat: align gpqa scoring with simple evals
|
1 tydzień temu |
orangels
|
898d02f7ef
|
feat: record upstream error debug payloads
|
1 tydzień temu |
orangels
|
5706ac8ee8
|
feat: add google gemini protocol
|
1 tydzień temu |
orangels
|
18cc7ff716
|
feat: add provider thinking parameters
|
1 tydzień temu |
orangels
|
ee13739c5e
|
feat: add raw response debug mode
|
1 tydzień temu |
orangels
|
e269f8f99d
|
docs: add end-to-end testing guide for RPM and benchmarks
|
1 tydzień temu |
orangels
|
f5b9239167
|
docs: add Chinese usage guide
|
1 tydzień temu |
orangels
|
f6b09d00a4
|
docs: plan rpm limiter modes
|
1 tydzień temu |
orangels
|
4a0e187e98
|
docs: specify benchmark report output
|
1 tydzień temu |
orangels
|
ea0b255547
|
docs: add implementation plan
|
1 tydzień temu |
orangels
|
61446dcf9f
|
docs: define benchmark dataset download flow
|
1 tydzień temu |
orangels
|
6e804b917f
|
docs: use official gsm8k benchmark
|
1 tydzień temu |
orangels
|
884097722f
|
docs: add design spec
|
1 tydzień temu |