GSO-Bench
General science and observation benchmark for frontier model capability tracking.
10rows
scoreprimary metric
2026-05-06sampled
Metadata
Metrics
Score, Standard error (lower is better)
| Rank | Subject | Score | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | GPT-5.2 | 27.40 |  GPT-5.2 openai-gpt-5.2 | Imported | 2026-05-06 |
| 2 | Claude Opus 4.5 | 26.50 |  Claude Opus 4.5 anthropic-claude-opus-4.5 | Imported | 2026-05-06 |
| 3 | Gemini 3 Pro | 18.60 |  Gemini 3 google-gemini-3 | Imported | 2026-05-06 |
| 4 | o3 | 8.80 | o3 openai-o3 | Imported | 2026-05-06 |
| 5 | kimi-k2-thinking (official) | 4.90 | MoonshotAI: Kimi K2 Thinking moonshotai-kimi-k2-thinking | Imported | 2026-05-06 |
| 6 | Qwen3-Max-Instruct | 4.90 |  Qwen3 Max qwen-qwen3-max | Imported | 2026-05-06 |
| 7 | Claude 3.7 Sonnet | 4.60 | Claude 3.7 Sonnet anthropic-claude-3.7-sonnet | Imported | 2026-05-06 |
| 8 | Gemini 2.5 Pro (Jun 2025) | 3.90 | Gemini 2.5 Pro google-gemini-2.5-pro | Imported | 2026-05-06 |
| 9 | o4-mini (high) | 3.60 | o4 Mini High openai-o4-mini-high | Imported | 2026-05-06 |
| 10 | GPT-4o | 0 | GPT-4o openai-gpt-4o | Imported | 2026-05-06 |
No matching rows.