Cell · cited by the thesis

complex_qwen3.6-35b_canon

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

Outcome
Wins (boss defeated, alive)0/5 [0.00, 0.43]
Final HP / exact optimum, mean over wins–
Died0
Final HP, mean143
Seconds per model call38.2
Parse error rate0.0%
Settings
FloorComplex floor
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
1not won167–3113replay
2not won166–3113replay
3not won173–3113replay
4not won173–3113replay
5not won34–3715replay
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>_complex_qwen3.6-35b_canon, never over the thesis result):

python scripts/run_catalog.py --codes VSC VSCa VSC5

Result file: results/reliability/complex_qwen3.6-35b_canon.json · updated 2026-10-03