ReLE - Agents - BFCL-V3: leaderboard
ReLE Chinese LLM benchmark (NoneLinear, formerly CLiB), a continuously updated Chinese-language evaluation run through each model's API. Agents sub-dimension: ReLE's own run of the Berkeley Function Calling Leaderboard V3 tasks; not a mirror of the Gorilla leaderboard. Accuracy 0-100, higher is better.
Metric: Accuracy (%). Source: nonelinear.com. Saturation forecast: Not forecast. 133 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | MiniMax-M2.7 | 76.5 | #270 |
| 2 | GLM-5.1 | 76 | #109 |
| 3 | Claude Opus 4.6 | 75.8 | #67 |
| 4 | GLM-5.2 | 75.7 | #80 |
| 5 | GLM-4.5 | 75.5 | #283 |
| 6 | Claude Opus 5.5 | 74.1 | #3 |
| 7 | Kimi K2 (Thinking) | 73.5 | #252 (Kimi K2) |
| 8 | Gemini 4 Argon | 72.5 | #9 |
| 9 | Qwen 3.5 Plus | 72.4 | #133 |
| 10 | GLM-4.6 | 72.3 | #265 |
| 11 | Step 3.7 Flash | 72.3 | #247 |
| 12 | GLM-5 | 71 | #149 |
| 13 | GLM 4.5 Air | 70.9 | #403 |
| 14 | Gemini 3.7 Flash | 70.8 | #24 |
| 15 | GLM-4.5 (Non-reasoning) | 70.8 | #283 (GLM-4.5) |
No result here: #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=rele-agents-bfcl-v3 · How It Works · Data refreshed daily, snapshot 2026-11-05.