Engram - Llama3.1-8b-Insturct - ALFWorld: leaderboard
Engram v1 A3.T6.6.2, native task performance (%), full reported method roster. Student backbone Llama3.1-8b-Insturct; trained adapters and memory frameworks remain native run configurations, not released-model aliases. Full table and confidence intervals retained; no resource, conditional verification or retention rate admitted.
Metric: Accuracy / task score (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama3.1-8b-Insturct (Engram Run: method=Engram;effort=source-unreported) | 83.6 | |
| 2 | Llama3.1-8b-Insturct (Engram Run: method=SkillRL;effort=source-unreported) | 80.6 | |
| 3 | Llama3.1-8b-Insturct (Engram Run: method=MAGE;effort=source-unreported) | 79.9 | |
| 4 | Llama3.1-8b-Insturct (Engram Run: method=EDGE;effort=source-unreported) | 78.4 | |
| 5 | Llama3.1-8b-Insturct (Engram Run: method=GRPO;effort=source-unreported) | 70.1 | |
| 6 | Llama3.1-8b-Insturct (Engram Run: method=EvolveR;effort=source-unreported) | 61.2 | |
| 7 | Llama3.1-8b-Insturct (Engram Run: method=Reflexion;effort=source-unreported) | 43.3 | |
| 8 | Llama3.1-8b-Insturct (Engram Run: method=MemRL;effort=source-unreported) | 36.6 | |
| 9 | Llama3.1-8b-Insturct (Engram Run: method=Mem0;effort=source-unreported) | 30.6 | |
| 10 | Llama3.1-8b-Insturct (Engram Run: method=ReAct;effort=source-unreported) | 25.4 |
Interactive version: theaggregate.ai/benchmark?slug=engram-llama3-1-8b-insturct-alfworld · How It Works · Data refreshed daily, snapshot 2026-11-12.