← older: fast__qwen3.6-35b-a3b__20260707-001626all runs · task boardnewer: fast2__qwen3.6-35b-a3b__20260707-111415 →

fast2qwen3.6-27b20260707-102314

No notes for this run yet. Add runs/fast2__qwen3.6-27b__20260707-102314/NOTES.md — what was run & what changed, results, interpretation, next steps — and it will render here.

Run details

modelllama-local/qwen3.6-27bagentharnesses.minimal_pi:MinimalPithinkingonreasoning budgetdirect (server/none)* — see journal for the authoritative mechanismmaxTokens / contextWindow65536 / 196608agent timeout ×2.0trials6 of 6 — 3 pass · 3 failmean reward0.50tokens (job total)5,409,583 in / 85,715 outstarted / finished2026-07-07T10:23 / 2026-07-07T11:14wall clock50m59s

Tasks

financial-document-processor — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	3m44s	2m56s	137507/6501	—	🔍 view

fix-ocaml-gc — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	22m13s	20m00s	2879436/35210	fast-timeout cut at 20mlong reasoning (33,434 chars) ×4	🔍 view

git-multibranch — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	3m44s	2m57s	408656/8584	—	🔍 view

headless-terminal — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	2m53s	2m00s	60257/4450	—	🔍 view

mailman — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	11m35s	9m24s	1699080/18279	claimed success but the verifier did NOT pass (heuristic)	🔍 view

reshard-c4-data — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	6m48s	5m00s	224647/12691	fast-timeout cut at 5mlong reasoning (16,634 chars)	🔍 view