← older: fast__qwen3.6-27b__20260706-175057all runs · task boardnewer: fast__qwen3.6-35b-a3b__20260706-201806 →

fastqwen3.6-27b20260706-184724

No notes for this run yet. Add runs/fast__qwen3.6-27b__20260706-184724/NOTES.md — what was run & what changed, results, interpretation, next steps — and it will render here.

Run details

modelllama-local/qwen3.6-27bagentharnesses.plan_pi:PlanPithinkingonreasoning budgetdirect (server/none)* — see journal for the authoritative mechanismmaxTokens / contextWindow65536 / 196608agent timeout ×2.0trials18 of 18 — 11 pass · 6 fail · 1 erroredmean reward0.61tokens (job total)3,236,501 in / 135,972 outstarted / finished2026-07-06T18:47 / 2026-07-06T20:18wall clock1h30m

Errored trials: VerifierTimeoutError (filter-js-from-html)

Tasks

cancel-async-tasks — 0/2 passed

#	result	total	agent	in/out tok	flags
1	FAIL	2m22s	1m33s	55209/4931	—	🔍 view
2	FAIL	2m35s	1m46s	98714/5399	—	🔍 view

filter-js-from-html — 0/2 passed

#	result	total	agent	in/out tok	flags
1	ERR	19m33s	4m00s	231335/12197	trial errored: VerifierTimeoutErrorfast-timeout cut at 4m	🔍 view
2	FAIL	6m54s	3m37s	173152/11673	claimed success but the verifier did NOT pass (heuristic)	🔍 view

fix-git — 2/2 passed

#	result	total	agent	in/out tok	flags
1	PASS	1m09s	32s	34753/1600	—	🔍 view
2	PASS	1m08s	31s	34660/1574	—	🔍 view

nginx-request-logging — 2/2 passed

#	result	total	agent	in/out tok	flags
1	PASS	1m23s	45s	59150/2368	—	🔍 view
2	PASS	1m26s	47s	48411/2493	—	🔍 view

openssl-selfsigned-cert — 2/2 passed

#	result	total	agent	in/out tok	flags
1	PASS	1m26s	48s	57737/2574	—	🔍 view
2	PASS	1m20s	44s	45546/2287	—	🔍 view

query-optimize — 0/2 passed

#	result	total	agent	in/out tok	flags
1	FAIL	10m45s	10m00s	831101/22049	a bash command timed outfast-timeout cut at 10m	🔍 view
2	FAIL	13m10s	6m31s	66718/3109	—	🔍 view

regex-log — 2/2 passed

#	result	total	agent	in/out tok	flags
1	PASS	6m56s	6m11s	244545/19370	long reasoning (20,816 chars)	🔍 view
2	PASS	6m18s	5m32s	287220/16545	long reasoning (20,156 chars)	🔍 view

sanitize-git-repo — 1/2 passed

#	result	total	agent	in/out tok	flags
1	PASS	3m00s	2m22s	398633/4612	—	🔍 view
2	FAIL	2m09s	1m31s	142754/3768	—	🔍 view

sparql-university — 2/2 passed

#	result	total	agent	in/out tok	flags
1	PASS	4m11s	3m24s	276661/9404	long reasoning (20,768 chars)	🔍 view
2	PASS	4m46s	4m00s	150202/10019	fast-timeout cut at 4mlong reasoning (26,118 chars)	🔍 view