Function Calling on BFCL (Berkeley Function Calling Leaderboard)
41.8Base ScoreGenEnv
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| GenEnvScale group=7B Models, Model Size=7B2025.12 | 41.8 | — | — | — | — | — | — | |
| Qwen 2.5 72BScale group=Large Scale Models (> 10B), Model Size=72B2025.12 | 35.3 | — | — | — | — | — | — | |
| Qwen 3 32BScale group=Large Scale Models (> 10B), Model Size=32B2025.12 | 33.8 | — | — | — | — | — | — | |
| Qwen 3 14BScale group=Large Scale Models (> 10B), Model Size=14B2025.12 | 29.4 | — | — | — | — | — | — | |
| GPT-OSS 20BScale group=Large Scale Models (> 10B), Model Size=20B2025.12 | 24.4 | — | — | — | — | — | — | |
| GPT-OSS 120BScale group=Large Scale Models (> 10B), Model Size=120B2025.12 | 21.9 | — | — | — | — | — | — | |
| Llama 3.1 70BScale group=Large Scale Models (> 10B), Model Size=70B2025.12 | 13.4 | — | — | — | — | — | — | |
| Qwen 2.5 7BScale group=7B Models, Model Size=7B2025.12 | 7 | — | — | — | — | — | — | |
| Llama 3.1 405BScale group=Large Scale Models (> 10B), Model Size=405B2025.12 | 5.5 | — | — | — | — | — | — | |
| ReSearchScale group=7B Models, Model Size=7B2025.12 | 5 | — | — | — | — | — | — | |
| SearchR1Scale group=7B Models, Model Size=7B2025.12 | 5 | — | — | — | — | — | — | |
| MatchTIR (KM)Backbone=Qwen3-8B2026.01 | 0.6 | 0.49 | 0.39 | 0.405 | 0.17 | 0.2602 | 0.3432 | |
| MatchTIR (OT)Backbone=Qwen3-8B2026.01 | 0.555 | 0.52 | 0.385 | 0.36 | 0.18 | 0.2387 | 0.3322 | |
| GRPOBackbone=Qwen3-8B2026.01 | 0.525 | 0.455 | 0.345 | 0.365 | 0.145 | 0.1892 | 0.2948 | |
| FTRL-SBackbone=Qwen3-8B2026.01 | 0.515 | 0.45 | 0.355 | 0.34 | 0.125 | 0.1742 | 0.2823 | |
| MatchTIR (KM)Backbone=Qwen3-4B2026.01 | 0.505 | 0.47 | 0.285 | 0.365 | 0.175 | 0.2215 | 0.3023 | |
| FTRL-MBackbone=Qwen3-8B2026.01 | 0.505 | 0.51 | 0.35 | 0.335 | 0.15 | 0.2194 | 0.3049 | |
| ToolRL-MBackbone=Qwen3-4B2026.01 | 0.5 | 0.375 | 0.29 | 0.3 | 0.105 | 0.228 | 0.2664 | |
| MatchTIR (OT)Backbone=Qwen3-4B2026.01 | 0.5 | 0.405 | 0.265 | 0.35 | 0.12 | 0.2172 | 0.2743 | |
| ToolRL-MBackbone=Qwen3-8B2026.01 | 0.5 | 0.47 | 0.395 | 0.32 | 0.155 | 0.1677 | 0.2913 | |
| VanillaBackbone=Qwen3-8B2026.01 | 0.475 | 0.46 | 0.375 | 0.345 | 0.1 | 0.2043 | 0.283 | |
| GRPOBackbone=Qwen3-4B2026.01 | 0.45 | 0.375 | 0.265 | 0.295 | 0.095 | 0.2194 | 0.2517 | |
| FTRL-MBackbone=Qwen3-4B2026.01 | 0.445 | 0.41 | 0.325 | 0.295 | 0.065 | 0.228 | 0.2576 | |
| FTRL-SBackbone=Qwen3-4B2026.01 | 0.43 | 0.355 | 0.31 | 0.28 | 0.065 | 0.2344 | 0.2467 | |
| VanillaBackbone=Qwen3-4B2026.01 | 0.415 | 0.31 | 0.265 | 0.275 | 0.09 | 0.1957 | 0.2296 | |
| ToolRL-SBackbone=Qwen3-8B2026.01 | 0.41 | 0.395 | 0.315 | 0.25 | 0.075 | 0.1763 | 0.2341 | |
| ToolRL-SBackbone=Qwen3-4B2026.01 | 0.325 | 0.31 | 0.225 | 0.2 | 0.075 | 0.1914 | 0.1991 | |
| ToRLScale group=7B Models, Model Size=7B2025.12 | 0 | — | — | — | — | — | — |