LifeAgentBench - Context Prompting: leaderboard
Metric: Answer Accuracy (%; single-user questions, pre-filtered records in the prompt, 32 new tokens). Source: arxiv.org. Saturation forecast: Around December 2026. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Pro | 58.89 |
| 2 | GPT-4o | 57.02 |
| 3 | Qwen 3.5 9B | 45.55 |
| 4 | Gemini 2.5 Flash Lite | 44.81 |
| 5 | Claude 3 Haiku | 35.3 |
| 6 | Gemma 2 9B (IT) | 24.44 |
| 7 | Llama 3.1 8B Instruct | 20.65 |
| 8 | Phi-3.5-mini-instruct | 20.57 |
| 9 | Llama 3.2 3B Instruct | 20.18 |
Interactive version: theaggregate.ai/benchmark?slug=lifeagentbench-context-prompting · How It Works · Data refreshed daily, snapshot 2026-09-25.