Agentic Task on AppWorld C (test)
48.4PR ScoreExpGraph
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ExpGraphExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM)2026.05 | 48.4 | 14.8 | |
| ReasoningBankExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 41.2 | 25.5 | |
| S3Executor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=LLM-Centric Experience Learning Baselines2026.05 | 40.8 | 22 | |
| Search-o1Executor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=LLM-Centric Experience Learning Baselines2026.05 | 39.8 | 24.6 | |
| ExpGraphExecutor LLM=Qwen3-32B (Small LLM)2026.05 | 39.3 | 13.6 | |
| IRCoTExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=LLM-Centric Experience Learning Baselines2026.05 | 36.1 | 17.8 | |
| S3Executor LLM=Qwen3-32B (Small LLM), Method Strategy Category=LLM-Centric Experience Learning Baselines2026.05 | 35 | 15 | |
| IRCoTExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=LLM-Centric Experience Learning Baselines2026.05 | 34.9 | 25.4 | |
| ReActExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=Prompt-based Agentic Baselines2026.05 | 32.1 | 16.1 | |
| MemRLExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 31.7 | 16.7 | |
| AWMExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 31.5 | 25.8 | |
| ReflexionExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=Prompt-based Agentic Baselines2026.05 | 31.1 | 15.6 | |
| ExpeLExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 29.9 | 28.3 | |
| LightMemExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 29.7 | 34.2 | |
| MemRLExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 28.9 | 24.7 | |
| ExpeLExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 28.2 | 25.6 | |
| Search-o1Executor LLM=Qwen3-32B (Small LLM), Method Strategy Category=LLM-Centric Experience Learning Baselines2026.05 | 26.5 | 14.2 | |
| LightMemExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 23.7 | 37.5 | |
| No-MemoryExecutor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=No-Memory2026.05 | 23.2 | 36.6 | |
| Mem0Executor LLM=Gemini-3.1-Flash-Lite (Large LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 22.9 | 34.8 | |
| AWMExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 22.8 | 33 | |
| ReasoningBankExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 22.2 | 24.3 | |
| ReflexionExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=Prompt-based Agentic Baselines2026.05 | 21.9 | 32.7 | |
| ReActExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=Prompt-based Agentic Baselines2026.05 | 21.8 | 31.9 | |
| No-MemoryExecutor LLM=Qwen3-32B (Small LLM), Method Strategy Category=No-Memory2026.05 | 21.6 | 31.5 | |
| Mem0Executor LLM=Qwen3-32B (Small LLM), Method Strategy Category=Retrieval-Centric Experience Learning Baselines2026.05 | 15.7 | 35.2 |