List your company

State Abstraction on AMA-Bench SA

52.92Success MetricOPD-Evolver-9B
27.512834.108940.70547.3011Jun 16, 2026

Evaluation results

MethodLinks
OPD-Evolver-9BBackbone LLM=QWEN3.5-9B2026.0652.92
QWEN3.5-397B-A17BBackbone LLM=QWEN3.5-397B-A17B2026.0651.41
STEP-3.5-FLASH (196B)Backbone LLM=STEP-3.5-FLASH (196B)2026.0648.88
AWMBackbone LLM=QWEN3.5-9B2026.0648
MemEvolveBackbone LLM=QWEN3.5-9B2026.0647.2
ReasoningBankBackbone LLM=QWEN3.5-9B2026.0647.04
MempBackbone LLM=QWEN3.5-9B2026.0646.59
CheatsheetBackbone LLM=QWEN3.5-9B2026.0646.19
EvolveRBackbone LLM=QWEN3.5-9B2026.0646.11
No MemoryBackbone LLM=QWEN3.5-9B2026.0645.63
ExpeLBackbone LLM=QWEN3.5-9B2026.0644.67
OPD-Evolver-4BBackbone LLM=QWEN3-4B2026.0634.9
AWMBackbone LLM=QWEN3-4B2026.0633.49
MemEvolveBackbone LLM=QWEN3-4B2026.0632.73
MempBackbone LLM=QWEN3-4B2026.0631.81
ReasoningBankBackbone LLM=QWEN3-4B2026.0630.97
CheatsheetBackbone LLM=QWEN3-4B2026.0630.37
EvolveRBackbone LLM=QWEN3-4B2026.0630.01
No MemoryBackbone LLM=QWEN3-4B2026.0629.73
ExpeLBackbone LLM=QWEN3-4B2026.0628.49