• Joined on 2026-07-22
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 17:15:39 +00:00
8c3d32bf2e swe scorer: return 'resolved' key (recipe metric name, not 'acc')
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 16:14:02 +00:00
9889fb6c0e swe scorer: unwrap grader's per-instance return shape
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 16:03:23 +00:00
7085624da0 swe_agentic: env_state carries version/hints/env_commit (make_test_spec need)
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 15:28:07 +00:00
014e8bcd9b swe scorer: metadata from env_state (positional pair misalign fix)
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 10:45:52 +00:00
4618c990b1 official swe scorer: PIP_INDEX_URL tuna in eval containers
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 10:42:31 +00:00
b3c837d3f3 swe_agentic: PIP_INDEX_URL env → tuna mirror in agent containers
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 10:35:40 +00:00
c7811fa774 swe_agentic: sentinel patch from container file, not text split
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 09:45:41 +00:00
fc8ade3c03 Background image service: pull-ahead + wait-barrier, unified lifecycle
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 09:03:19 +00:00
e9e3230e46 swe_agentic: auto-release pulled images at process exit
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 08:54:31 +00:00
f09fe184fe swe_agentic: ensure per-instance image before docker run
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 08:23:02 +00:00
a93996094d swe_agentic: sentinel patch must end with a newline
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 08:11:20 +00:00
565ba11790 swe datasets: carry version/patch/hints/created_at in metadata
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 04:57:40 +00:00
ace38eaf76 thinking.yaml: swe_bench_verified_agentic entry (env + max_turns)
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 04:01:50 +00:00
2a2335f842 swe_agentic: accept adapter kwarg in __init__ (get_env uniform signature)
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 02:41:00 +00:00
de847206e5 swe scoring: official swebench harness (es-identical path), backend hook
sora pushed to main at Meta-Eval/EvalHarness 2026-09-18 02:29:37 +00:00
f662006517 swe_bench_verified_agentic: multi-turn SWE agent (mini-swe-agent protocol)
sora pushed to main at Meta-Eval/EvalHarness 2026-09-17 10:10:06 +00:00
f4e2c2d1b4 config: swe_bench_verified entry (patch generation params)
sora pushed to main at Meta-Eval/EvalHarness 2026-09-17 09:38:19 +00:00
c23665fed7 Scoring preflight: auto batch-pull for multi-image benches
sora pushed to main at Meta-Eval/EvalHarness 2026-09-17 09:36:24 +00:00
587274d8b6 sandbox pull: SWE-bench per-instance pre-pull with resume (es port)
sora pushed to main at Meta-Eval/EvalHarness 2026-09-17 09:24:55 +00:00
f844775f1e tau2 scorer: read the composite 'reward' field