Cell · cited by the thesis

hand02_qwen3.6-35b_canon

Qwen3.6-35B on hand02, 5 episodes. Cited by the thesis in:

Outcome
Wins (boss defeated, alive)3/5 [0.23, 0.88]
Final HP / exact optimum, mean over wins31%
Died0
Final HP, mean21
Seconds per model call33.5
Parse error rate0.0%
Settings
Floorhand02
ModelQwen3.6-35B (qwen3.6-35b-a3b, thinking)
Planningreactive
Promptengine-default-floor-basic
Temperature0
Lookahead depth3
Mechanical overrideson (logged per turn)
Engine settings (environment, 5)
THINKING_MAX_TOKENS=16384
CONVERSATION_HISTORY_TURNS=0
MOTA_ORDER_TWO_SIDED=1
MOTA_FIGHT_KEY_FLOOD=1
LLM_TIMEOUT=300
Episodesopen one to replay it turn by turn
#outcomefinal HPHP / optstepsmodel callsreplay
1won431%279replay
2won431%279replay
3not won67–3113replay
4not won24–3113replay
5won431%279replay
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>_hand02_qwen3.6-35b_canon, never over the thesis result):

python scripts/run_catalog.py --codes VS2 VS2a VS25

Result file: results/reliability/hand02_qwen3.6-35b_canon.json · updated 2026-10-03