EvalHarness

History

sora 2dfb60cdf2 Few-shot split loading goes through cached materialization

Was: raw load_raw_records() hit the hub on EVERY run -- offline machines
stalled in 5x HF retries then silently degraded to 0-shot (changing the
benchmark's default contract, e.g. gsm8k 4-shot). Now the few-shot split
is a proper Dataset entry: first use downloads+cache, every later run is
a pure cache hit (verified: second run with HF_HUB_OFFLINE=1 loads
4-shot from cache, zero network lines, acc unchanged).

Co-Authored-By: Claude <noreply@anthropic.com>

2026-09-10 07:47:48 +00:00

__init__.py

Add model layer: async ModelAdapter (openai_compatible + mock) returning structured ModelOutput(text,tool_calls,usage), Deployer registry (vllm/sglang docker-pinned via models.yaml, external), async run_eval generate->score, CLI --model, agent-ready SampleResult.trajectory/env_state, tests

2026-08-24 06:28:00 +00:00

adapter.py

dp4-flash 28-bench alignment: renderer plugin layer, exec_workers, gen_profiles, AIMD pool, BCB/LCB/bfcl/gfc judge fixes

2026-09-08 05:55:04 +00:00

deployer.py

Add sandbox layer (docker exec hard-isolation + serve envs, refcounted acquire/release, atexit teardown, bind-mount sharing) and agent evaluation driver (message pump drive(), Trajectory, bfcl_mock env with official call-sequence scoring, mock:fc oracle); Deployer delegates to sandbox.serve_env; code_any extractor fixes indentation-stripping; CLI --env

2026-08-24 07:11:03 +00:00

gen_profiles.py

dp4-flash 28-bench alignment: renderer plugin layer, exec_workers, gen_profiles, AIMD pool, BCB/LCB/bfcl/gfc judge fixes