Letta Memory Benchmark - Core Memory Write (Hard): leaderboard
The Letta agent must write information into its core memory block through the memory tool, on the task's hard variant, over 100 items repeated 3 times; the score is the share done correctly.
Metric: Score (%) on 100 hard core-memory write items, 3 repeats, Letta agent framework. Source: raw.githubusercontent.com. Saturation forecast: Estimated already saturated. 18 models tracked. Updated 16 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Sonnet 4 (20250514) | 100 | #388 | |
| 2 | Claude Sonnet 4 | Thinking | 100 | #406 |
| 3 | GPT-4o Mini (2024-07-18) | 98.33 | #1511 | |
| 4 | Gemini 2.5 Flash (Preview 05-20) | 97 | #423 | |
| 5 | GPT-4o (2024-08-06) | 96.67 | #736 | |
| 6 | Gemini 2.5 Flash (Preview 04-17) | 95.67 | #458 | |
| 7 | Gemini 2.5 Pro (Preview 05-06) | 94.33 | #279 | |
| 8 | O4 Mini | 93.33 | #356 | |
| 9 | Claude 3.7 Sonnet (20250219) | 92 | #414 | |
| 10 | Claude 3.5 Haiku (20241022) | 91.67 | #1601 | |
| 11 | GPT-4.1 Nano | 78.67 | #2035 | |
| 12 | Llama 4 Scout Instruct | 78.67 | #1579 | |
| 13 | Llama 3.3 70B Instruct | 73 | #1379 | |
| 14 | GPT-4.1 Mini | 70 | #781 | |
| 15 | Llama 4 Maverick Instruct FP8 | 56.67 | #1010 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=letta-memory-benchmark-core-memory-write-hard · How It Works · Data refreshed daily, snapshot 2026-11-16.