Skip to main content

BFCL v4 Leaderboard — AI Model Benchmarks — MCP App Store | MCP App Store

AI Model Leaderboard

Compare AI models across agentic, reasoning, coding, and tool-use benchmarks.

Tool Use

BFCL v4 Tool-Decathlon

Agentic Reasoning Coding Tool Use Computer Use

Benchmarks

agentic

reasoning

coding

tool-use

computer-use

BFCL v4

Berkeley Function-Calling Leaderboard v4. Tests tool-use accuracy across serial, parallel, multi-turn, and agentic interactions with web search.

Tool Use44 models · % accuracy

MCP App Store

The AI ecosystem directory — MCP Apps, Agent Skills, and daily news.

Directory

MCP Apps & Servers Agent Skills AI News AI Models What are MCP Apps?

Resources

Documentation Specification GitHub

Account

Sign in Get Started Dashboard

Company

Contact Build an MCP App

Legal

Privacy Policy Terms of Use

© 2026 MCP App Store

All rights reserved.

Claude Opus 4.6Undisclosed

77.47% accuracy

Claude Sonnet 4.6Undisclosed

73.24% accuracy

72.80% accuracy

Gemini 3 ProUndisclosed

72.51% accuracy

72.38% accuracy

Grok 4.1 FastUndisclosed

69.57% accuracy

Claude Haiku 4.5Undisclosed

68.70% accuracy

63.05% accuracy

Grok 4Undisclosed

62.97% accuracy

Kimi-K2-Instruct1T

59.06% accuracy

DeepSeek-V3.2671B

56.73% accuracy

DeepSeek-V3.2-Exp671B

56.73% accuracy

Gemini 2.5 FlashUndisclosed

56.24% accuracy

GPT-5.4Undisclosed

55.87% accuracy

GPT-5.4 MiniUndisclosed

55.46% accuracy

GPT-4.1Undisclosed

53.96% accuracy

o4-miniUndisclosed

53.24% accuracy

Qwen3-235B-A22B235B

52.15% accuracy

GPT-5.4 NanoUndisclosed

51.45% accuracy

Nanbeige4.1-3B3B

51.40% accuracy

GPT-4.1 MiniUndisclosed

50.45% accuracy

48.71% accuracy

48.71% accuracy

42.57% accuracy

Qwen2.5-14B-Instruct14B

42.57% accuracy