MLX Benchmark V2 - MLX Embeddings LoRA — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 22 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 Nano | 86.67 |
| 2 | Claude Sonnet 4.6 | 66.67 |
| 3 | Gemma 4 26B A4B (IT) | 60 |
| 4 | Gemini 3 Flash (Preview) | 60 |
| 5 | Gemini 2.5 Flash Lite (Preview 09-2025) | 60 |
| 6 | DeepSeek V4 Flash | 53.33 |
| 7 | Grok 4.1 Fast | 53.33 |
| 8 | GPT-5 Nano | 46.67 |
| 9 | Qwen 3.6 35B A3B | 33.33 |
| 10 | Qwen 3.6 Max Preview | 33.33 |
| 11 | DeepSeek V4 Pro | 30.77 |
| 12 | Kimi K2.5 | 6.67 |
| 13 | GLM-5.1 | 6.67 |
| 14 | Kimi K2.6 | 6.67 |
| 15 | Nemotron 3 Ultra | 6.67 |
Interactive version: theaggregate.ai/benchmark?slug=mlx-benchmark-v2-mlx-embeddings-lora · How the rankings work · Data refreshed daily, snapshot 2026-07-22.