Cell

hand03_qwen3.6-35b_final2

Qwen3.6-35B on hand03, 3 episodes. Not cited by the thesis (development or earlier run).

Outcome
Wins (boss defeated, alive)0/3 [0.00, 0.56]
Final HP / exact optimum, mean over wins–
Died0
Final HP, mean59
Seconds per model call24.8
Parse error rate0.0%
Settings
Floorhand03
ModelQwen3.6-35B (qwen3.6-35b-a3b, thinking)
Planningreactive
Promptengine-default-floor-basic
Temperature0
Lookahead depth3
Mechanical overrideson (logged per turn)
Engine settings (environment, 4)
THINKING_MAX_TOKENS=16384
CONVERSATION_HISTORY_TURNS=0
MOTA_ORDER_TWO_SIDED=1
MOTA_FIGHT_KEY_FLOOD=1
Episodesopen one to replay it turn by turn
#outcomefinal HPHP / optstepsmodel callsreplay
1not won4–3614replay
2not won86–3614replay
3not won86–3614replay
Re-run this cell

In a checkout of mt-ai-lab with an API key set (the result goes to a new file rerun_<time>_hand03_qwen3.6-35b_final2, never over the thesis result):

python scripts/run_catalog.py --codes HS3

Result file: results/reliability/hand03_qwen3.6-35b_final2.json · updated 2026-09-29