List your company

Tool-use on BFCL Multi-turn

54.75AccuracyClaude-Sonnet-4
21.958830.471938.98547.4981Dec 22, 2025Dec 29, 2025Jan 5, 2026Jan 12, 2026Jan 19, 2026Jan 26, 2026Feb 3, 2026

Evaluation results

MethodLinks
Claude-Sonnet-4Category=Proprietary Models2026.0254.7555.91————
AWPOParameter=4B2025.1252.12—59593951.5
DeepSeek-V3.2Category=Open-Source Foundation Models2026.0244.8869.75————
DeepSeek-R1-0528Parameter=671B2025.1244.5—54.54136.546
GPT-5Category=Proprietary Models2026.0243.7568.69————
Moonshotai-Kimi-K2-InstParameter=1000B2025.1241.25—51433140
TDScalingBase Model=Qwen3-Coder-30B-A3B-Instruct, Training Samples=50002026.0240.4452.47————
Qwen3-235B-A22BParameter=235B2025.1240.12—494129.541
GPT-4.1Category=Proprietary Models2026.0238.8850.72————
watt-tool-8BParameter=8B2025.1237.88—45.5392443
BitAgent-8BParameter=8B2025.1237.75—46.537.52443
TOUCANCategory=Tool-Learning Methods, Training Samples=50002026.0237.0345.92————
ToolACE-2-8BParameter=8B2025.1237—47312842
TDScalingBase Model=Qwen3-Coder-30B-A3B-Instruct, Training Samples=5002026.0236.6648.99————
TDScalingBase Model=Qwen3-30B-A3B-Instruct, Training Samples=5002026.0236.6341.5————
Grok-4-0709Parameter=/2025.1236.12—44312643.5
Qwen3-Coder-480B-A35B-InstructCategory=Open-Source Foundation Models2026.0235.9151.85————
Qwen3-30B-A3B-InstructCategory=Base Model2026.0233.2234.74————
Qwen3-Coder-30B-A3B-InstructCategory=Base Model2026.0229.4141.48————
Gemini-2.5-proCategory=Proprietary Models2026.0229.2544.99————
GPT-5-2025-08-07Parameter=/2025.1228.5—33.529.52328
APIGen-MTCategory=Tool-Learning Methods, Training Samples=50002026.0227.2542.81————
Gemini-2.5-ProParameter=/2025.1225—25.52624.524
SimiaCategory=Tool-Learning Methods, Training Samples=50002026.0223.2245.29————