← older: smoke__qwen3.6-35b-a3b__20260726-042939all runs · task boardnewer: smoke__laguna-s-2.1__20260726-193806 →

smokeqwen3.6-35b-a3b20260726-100656

No notes for this run yet. Add runs/smoke__qwen3.6-35b-a3b__20260726-100656/NOTES.md — what was run & what changed, results, interpretation, next steps — and it will render here.

Run details

modelllama-local/qwen3.6-35b-a3bagentharnesses.minimal_pi:MinimalPithinkingonreasoning budgetdirect (server/none)* — see journal for the authoritative mechanismmaxTokens / contextWindow65536 / 196608agent timeout ×2.0trials39 of 39 — 23 pass · 16 failmean reward0.59tokens (job total)96,706,714 in / 1,685,000 outstarted / finished2026-07-26T10:06 / 2026-07-26T16:18wall clock6h11m

Tasks

bn-fit-modify — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	5m30s	2m49s	187385/8692	—	🔍 view

break-filter-js-from-html — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	20m34s	20m00s	1330775/141046	fast-timeout cut at 20mlong reasoning (29,065 chars) ×13loop-guard blocked a repeated call	🔍 view

build-cython-ext — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	3m52s	2m58s	2322497/10098	—	🔍 view

build-pmars — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	1m39s	1m03s	251327/5882	—	🔍 view

cancel-async-tasks — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	5m03s	3m36s	504862/31328	long reasoning (21,953 chars) ×2	🔍 view

cobol-modernization — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	6m41s	6m08s	4539671/55678	long reasoning (12,070 chars) ×2	🔍 view

code-from-image — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	3m07s	1m55s	131714/12434	—	🔍 view

count-dataset-tokens — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	2m06s	1m32s	133516/5501	—	🔍 view

crack-7z-hash — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	2m52s	2m12s	113662/2872	—	🔍 view

financial-document-processor — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	7m48s	5m25s	694247/23960	claimed success but the verifier did NOT pass (heuristic)	🔍 view

fix-git — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	52s	18s	75437/2451	—	🔍 view

fix-ocaml-gc — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	31m29s	7m43s	2767312/21708	long reasoning (25,114 chars) ×2	🔍 view

git-multibranch — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	3m10s	2m27s	1061739/16461	—	🔍 view

headless-terminal — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	2m29s	1m39s	163753/9687	a bash command timed out	🔍 view

kv-store-grpc — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	56s	24s	78337/2645	—	🔍 view

large-scale-text-editing — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	18m37s	17m21s	5460805/110084	a bash command timed out ×5long reasoning (20,865 chars) ×9	🔍 view

largest-eigenval — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	12m21s	11m50s	11416684/95815	long reasoning (12,019 chars) ×6	🔍 view

llm-inference-batching-scheduler — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	20m34s	20m00s	1528976/80245	a bash command timed outfast-timeout cut at 20mlong reasoning (20,560 chars) ×4	🔍 view

mailman — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	20m53s	19m16s	18398382/62714	a bash command timed out ×2long reasoning (12,650 chars) ×2	🔍 view

merge-diff-arc-agi-task — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	2m35s	1m52s	371298/18902	long reasoning (12,051 chars) ×2	🔍 view

nginx-request-logging — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	53s	15s	40759/2220	—	🔍 view

openssl-selfsigned-cert — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	47s	13s	28794/2088	—	🔍 view

overfull-hbox — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	20m58s	20m00s	13551790/160314	fast-timeout cut at 20mlong reasoning (23,638 chars) ×11loop-guard blocked a repeated call	🔍 view

pypi-server — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	1m11s	37s	111967/3431	—	🔍 view

pytorch-model-cli — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	7m22s	4m03s	1150280/28301	—	🔍 view

pytorch-model-recovery — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	10m05s	4m48s	307233/16072	a bash command timed out	🔍 view

qemu-alpine-ssh — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	20m35s	20m00s	2525694/50748	a bash command timed out ×4fast-timeout cut at 20m	🔍 view

qemu-startup — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	20m39s	20m00s	6501491/81262	a bash command timed out ×2fast-timeout cut at 20mlong reasoning (29,060 chars) ×5	🔍 view

query-optimize — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	14m45s	7m09s	481472/12244	a bash command timed out ×2	🔍 view

regex-log — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	21m46s	20m00s	3148241/169289	fast-timeout cut at 20mlong reasoning (18,872 chars) ×10	🔍 view

reshard-c4-data — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	6m56s	5m23s	3227731/34384	long reasoning (25,555 chars)runaway / empty-final recovery fired	🔍 view

rstan-to-pystan — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	10m28s	9m46s	1297284/23228	long reasoning (23,349 chars)	🔍 view

sanitize-git-repo — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	1m55s	1m21s	943250/9898	—	🔍 view

sparql-university — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	4m38s	3m48s	1826181/26174	long reasoning (23,923 chars)	🔍 view

sqlite-db-truncate — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	13m28s	12m54s	1516025/111541	long reasoning (15,001 chars) ×10	🔍 view

sqlite-with-gcov — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	1m26s	42s	89046/2256	—	🔍 view

torch-tensor-parallelism — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	16m22s	10m43s	2419685/46507	long reasoning (32,597 chars) ×3	🔍 view

tune-mjcf — 0/1 passed

#	result	total	agent	in/out tok	flags
1	FAIL	19m22s	18m35s	5587476/151159	loop-guard hard-stopped the sessionlong reasoning (31,954 chars) ×6loop-guard blocked a repeated call ×11loop-guard escalation nudge	🔍 view

vulnerable-secret — 1/1 passed

#	result	total	agent	in/out tok	flags
1	PASS	4m15s	3m40s	419936/35681	long reasoning (13,965 chars) ×2	🔍 view