SEAL - Agentic Tool Use (Chat): leaderboard
Scale SEAL evaluation of tool use in conversational settings, testing ability to select and chain tools for user requests.
Metric: Score. Source: scale.com. Saturation forecast: Estimated already saturated. 35 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 Mini (High) | 63.45 | #272 (O3 Mini) |
| 2 | Gemini 2.5 Pro (Preview 03-25) | 62.43 | #123 |
| 3 | O3 Mini (Medium) | 62.42 | #272 (O3 Mini) |
| 4 | O1 Pro | 61.42 | #228 |
| 5 | DeepSeek R1 | 60.91 | #251 |
| 6 | O1 (2024-12-17) | 60.41 | #149 |
| 7 | DeepSeek V3 | 58.55 | #321 |
| 8 | Gemini 2.0 Pro (Preview 02-05) | 57.86 | #307 |
| 9 | Gemini 2.0 Flash (Thinking) | 57.36 | #342 (Gemini 2.0 Flash) |
| 10 | GPT-4o (2024-08-06) | 56.85 | #336 |
| 11 | GPT-4.5 (Preview) | 56.34 | #244 |
| 12 | Claude 3.7 Sonnet (20250219) | 56.25 | #201 |
| 13 | Claude 3.5 Sonnet (20240620) | 56.06 | #299 |
| 14 | Claude 3.7 Sonnet (Thinking) | 55.32 | #249 (Claude 3.7 Sonnet) |
| 15 | O1 Preview | 55.1 | #193 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=seal-agentic-tool-use-chat · How It Works · Data refreshed daily, snapshot 2026-10-20.