← older: smoke__qwen3.6-35b-a3b__20260706-092006all runs · task boardnewer: fast__qwen3.6-35b-a3b__20260706-113104

smoke__qwen3.6-35b-a3b__20260706-093016

No notes for this run yet. Add runs/smoke__qwen3.6-35b-a3b__20260706-093016/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 ×2.0trials20 of 20 — 17 pass · 3 failmean reward0.85tokens (job total)4,426,033 in / 291,006 outstarted / finished2026-07-06T09:30 / 2026-07-06T10:14wall clock44m28s

Tasks

fix-git — 4/5 passed

#resulttotalagentin/out tokflags
1PASS45s9s33201/1196
🔍 view
2FAIL1m35s57s173586/9511
🔍 view
3PASS47s11s34643/1641
🔍 view
4PASS46s9s34601/1240
🔍 view
5PASS52s15s60079/2120
🔍 view

nginx-request-logging — 5/5 passed

#resulttotalagentin/out tokflags
1PASS54s17s37177/2441
🔍 view
2PASS54s17s44052/2430
🔍 view
3PASS53s15s43907/2286
🔍 view
4PASS55s18s62977/2425
🔍 view
5PASS51s13s50326/1984
🔍 view

openssl-selfsigned-cert — 3/5 passed

#resulttotalagentin/out tokflags
1FAIL58s22s54548/3431
🔍 view
2PASS1m00s22s83900/3635
🔍 view
3PASS53s18s31472/2963
🔍 view
4PASS52s16s43620/2696
🔍 view
5FAIL57s21s61286/3112
🔍 view

regex-log — 5/5 passed

#resulttotalagentin/out tokflags
1PASS8m17s7m33s1207111/68940
long reasoning (19,170 chars) ×4
🔍 view
2PASS8m26s7m43s974428/69016
long reasoning (19,867 chars) ×6
🔍 view
3PASS5m33s4m49s476672/45446
long reasoning (23,690 chars) ×4
🔍 view
4PASS2m34s1m51s343340/18203
long reasoning (19,893 chars)
🔍 view
5PASS5m35s4m51s575107/46290
long reasoning (21,402 chars) ×4
🔍 view