← BFCL leaderboards

Memory Recursive Summarization

BFCL · 50 models

RankModelScore
1Claude-Sonnet-4-5-20250929 (FC)83.23
2Claude-Opus-4-5-20251101 (FC)77.42
3o3-2025-04-16 (FC)72.9
4o3-2025-04-16 (Prompt)71.61
5GLM-4.6 (FC thinking)67.1
6Gemini-3-Pro-Preview (Prompt)63.23
6Grok-4-1-fast-reasoning (FC)63.23
8GPT-5.2-2025-12-11 (FC)60.65
9DeepSeek-V3.2-Exp (FC)59.35
10o4-mini-2025-04-16 (FC)58.71
11o4-mini-2025-04-16 (Prompt)58.06
12Claude-Haiku-4-5-20251001 (FC)56.13
13Gemini-2.5-Flash (Prompt)55.48
14Gemini-2.5-Flash (FC)54.19
15GPT-5-mini-2025-08-07 (FC)52.26
16Gemini-2.5-Flash-Lite (FC)51.61
16Grok-4-0709 (FC)51.61
18Falcon3-10B-Instruct (FC)50.97
18Gemini-3-Pro-Preview (FC)50.97
20Gemma-3-12b-it (Prompt)49.03
21Grok-4-0709 (Prompt)48.39
22Command A Reasoning (FC)46.45
23Qwen3-235B-A22B-Instruct-2507 (FC)45.81
23xLAM-2-32b-fc-r (FC)45.81
25Moonshotai-Kimi-K2-Instruct (FC)45.16
26Nanbeige4-3B-Thinking-2511 (FC)44.52
26Qwen3-4B-Instruct-2507 (Prompt)44.52
28GPT-4.1-mini-2025-04-14 (FC)41.94
28Qwen3-32B (FC)41.94
30Command A (FC)40
31DeepSeek-V3.2-Exp (Prompt + Thinking)39.35
31Nanbeige3.5-Pro-Thinking (FC)39.35
33Falcon3-7B-Instruct (FC)38.71
33GPT-4.1-mini-2025-04-14 (Prompt)38.71
33GPT-5-mini-2025-08-07 (Prompt)38.71
36Grok-4-1-fast-non-reasoning (FC)38.06
37GPT-4.1-2025-04-14 (FC)37.42
38GPT-4.1-2025-04-14 (Prompt)35.48
38Gemma-3-27b-it (Prompt)35.48
40Qwen3-14B (FC)34.84
40Qwen3-30B-A3B-Instruct-2507 (FC)34.84
42Mistral-Medium-250534.19
43Mistral-Medium-2505 (FC)33.55
43Qwen3-235B-A22B-Instruct-2507 (Prompt)33.55
43ToolACE-2-8B (FC)33.55
46Mistral-small-2506 (FC)31.61
46Phi-4 (Prompt)31.61
46Qwen3-8B (FC)31.61
49Mistral-Small-2506 (Prompt)30.97
50Gemini-2.5-Flash-Lite (Prompt)29.68