Python Parallel Multiple AST

BFCL · 50 models

Rank	Model	Score
1	BitAgent-Bounty-8B	95.83
2	Gemini-3-Pro-Preview (Prompt)	87.5
3	Claude-Sonnet-4-5-20250929 (FC)	83.33
4	Amazon-Nova-2-Lite-v1:0 (FC)	79.17
4	Arch-Agent-32B	79.17
4	Gemini-3-Pro-Preview (FC)	79.17
4	Grok-4-0709 (FC)	79.17
4	Llama-4-Scout-17B-16E-Instruct (FC)	79.17
4	Qwen3-8B (FC)	79.17
10	Arch-Agent-3B	75
10	Claude-Haiku-4-5-20251001 (FC)	75
10	Claude-Opus-4-5-20251101 (FC)	75
10	Command A Reasoning (FC)	75
10	GLM-4.6 (FC thinking)	75
10	GPT-4.1-2025-04-14 (Prompt)	75
10	Gemini-2.5-Flash (Prompt)	75
10	Hammer2.1-7b (FC)	75
10	Qwen3-1.7B (FC)	75
10	Qwen3-14B (Prompt)	75
10	Qwen3-30B-A3B-Instruct-2507 (FC)	75
10	Qwen3-8B (Prompt)	75
10	mistral-large-2411 (FC)	75
23	Amazon-Nova-Pro-v1:0 (FC)	70.83
23	Command A (FC)	70.83
23	Command R7B (FC)	70.83
23	GPT-4.1-2025-04-14 (FC)	70.83
23	GPT-4.1-mini-2025-04-14 (Prompt)	70.83
23	Gemini-2.5-Flash (FC)	70.83
23	Grok-4-0709 (Prompt)	70.83
23	Grok-4-1-fast-non-reasoning (FC)	70.83
23	Grok-4-1-fast-reasoning (FC)	70.83
23	Llama-3.3-70B-Instruct (FC)	70.83
23	Llama-4-Maverick-17B-128E-Instruct-FP8 (FC)	70.83
23	Mistral-small-2506 (FC)	70.83
23	Nanbeige4-3B-Thinking-2511 (FC)	70.83
23	Qwen3-14B (FC)	70.83
23	Qwen3-235B-A22B-Instruct-2507 (Prompt)	70.83
23	o3-2025-04-16 (Prompt)	70.83
23	palmyra-x-004 (FC)	70.83
40	Gemini-2.5-Flash-Lite (FC)	66.67
40	Hammer2.1-3b (FC)	66.67
40	Mistral-Small-2506 (Prompt)	66.67
40	Moonshotai-Kimi-K2-Instruct (FC)	66.67
40	Nanbeige3.5-Pro-Thinking (FC)	66.67
40	Open-Mistral-Nemo-2407 (FC)	66.67
40	Qwen3-30B-A3B-Instruct-2507 (Prompt)	66.67
40	Qwen3-4B-Instruct-2507 (FC)	66.67
48	Claude-Opus-4-5-20251101 (Prompt)	62.5
48	CoALM-70B	62.5
48	Falcon3-7B-Instruct (FC)	62.5