SAGE - Linear Algebra
Math · 2025-10-08
The Linear Algebra split of SAGE. This child benchmark separates a source-reported subtask or subtrack from the parent aggregate so scores at different grains do not share one benchmark_slug.
Top models (higher is better)
| Model | Score |
|---|---|
| Opus 4.8 | 78.3 |
| Opus 4.7 | 77.0 |
| Claude Opus 5 | 76.1 |
| Opus 4.6 | 75.4 |
| GPT-5.5 | 74.4 |
| Gemini 3.1 Pro Preview | 74.2 |
| Kimi K3 | 72.7 |
| Gemma 4 31B IT | 72.6 |
| Claude Fable 5 | 71.5 |
| GPT-5.4 Mini | 71.5 |
| Gemini 3 Flash Preview | 71.0 |
| Kimi K2.5 | 71.0 |
| Opus 4.5 | 70.6 |
| GPT-5.6 Sol | 70.4 |
| Gemini 3.6 Flash | 69.9 |
| GPT-5.2 | 69.6 |
| Sonnet 5 | 69.0 |
| Muse Spark 1.1 | 68.9 |
| GPT-5.4 | 68.8 |
| GPT-5.1 | 68.6 |
| Gemini 3.5 Flash | 68.3 |
| Kimi K2.6 | 68.1 |
| MiniMax M3 | 66.8 |
| Gemini 3 Pro Preview | 65.1 |
| GPT-5.6 Terra | 64.0 |
| Llama 4 Maverick Instruct | 63.9 |
| Gemini 3.5 Flash-Lite | 63.7 |
| GPT-5.6 Luna | 63.2 |
| Sonnet 4.6 | 63.2 |
| GPT-5 Mini | 62.7 |
| GPT-5 | 61.6 |
| o3 | 61.4 |
| Llama 4 Scout Instruct | 61.2 |
| GPT-5.4 Nano | 60.7 |
| Qwen3.6 Plus (2026-04-02) | 60.0 |
| Qwen3.6 27B | 59.7 |
| Gemini 3.1 Flash-Lite Preview | 58.9 |
| MiMo-V2.5 | 58.8 |
| Gemini 2.5 Flash | 58.5 |
| Gemini 2.5 Flash Preview (09-2025) | 57.3 |