MLX Benchmark V2 - True/False — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2.5 | 100 |
| 2 | GLM-5.1 | 100 |
| 3 | DeepSeek V4 Flash | 100 |
| 4 | Gemini 2.5 Flash Lite (Preview 09-2025) | 100 |
| 5 | Claude Sonnet 4.6 | 91.67 |
| 6 | DeepSeek V4 Pro | 91.67 |
| 7 | Gemini 3 Flash (Preview) | 91.67 |
| 8 | GPT-5.4 Nano | 91.67 |
| 9 | Qwen 3.6 Max Preview | 91.67 |
| 10 | Gemma 4 26B A4B (IT) | 83.33 |
| 11 | Qwen 3.6 35B A3B | 83.33 |
| 12 | Grok 4.1 Fast | 83.33 |
| 13 | Nemotron 3 Ultra | 83.33 |
| 14 | GPT-5 Nano | 75 |
| 15 | Kimi K2.6 | 66.67 |
Interactive version: theaggregate.ai/benchmark?slug=mlx-benchmark-v2-true-false · How the rankings work · Data refreshed daily, snapshot 2026-07-22.