tower_deepseek-v4-flash_final2_plan2
DeepSeek-V4-Flash on the tower (MT0→MT23w), 1 run. Not cited by the thesis (development or earlier run).
| Highest floor reached, per run | MT7 |
| Died | 0 |
| Final HP, mean | 1943 |
| Seconds per model call | 6.6 |
| Parse error rate | 2.9% |
| Floor | Tower MT0→MT23w |
| Model | DeepSeek-V4-Flash (deepseek-v4-flash-0731, thinking) |
| Planning | reactive |
| Prompt | engine-default-tower-basic |
| Temperature | 0 |
| Lookahead depth | 2 |
| Mechanical overrides | on (logged per turn) |
Engine settings (environment, 6)
THINKING_MAX_TOKENS=16384 CONVERSATION_HISTORY_TURNS=0 MOTA_ORDER_TWO_SIDED=1 MOTA_FIGHT_KEY_FLOOD=1 MOTA_PLAN_STEPS=2 LLM_TIMEOUT=1800
| # | outcome | final HP | HP / opt | turns | model calls | replay |
|---|---|---|---|---|---|---|
| 1 | MT7 | 1943 | – | 192 | 102 | replay |
In a checkout of mt-ai-lab with an API key set (the result goes to a new file rerun_<time>_tower_deepseek-v4-flash_final2_plan2, never over the thesis result):
python scripts/run_catalog.py --codes DTP2
Result file: results/reliability/tower_deepseek-v4-flash_final2_plan2.json · updated 2026-09-30