Non-Live Parallel AST

BFCL · 50 models

Rank	Model	Score
1	Arch-Agent-32B	96
1	Gemini-2.5-Flash (Prompt)	96
3	Qwen3-14B (Prompt)	95.5
3	Qwen3-235B-A22B-Instruct-2507 (Prompt)	95.5
5	Gemini-3-Pro-Preview (Prompt)	95
6	Claude-Sonnet-4-5-20250929 (FC)	94.5
6	Qwen3-8B (Prompt)	94.5
8	GPT-4.1-2025-04-14 (Prompt)	94
8	Gemini-2.5-Flash (FC)	94
8	Llama-4-Maverick-17B-128E-Instruct-FP8 (FC)	94
8	Llama-4-Scout-17B-16E-Instruct (FC)	94
8	Qwen3-30B-A3B-Instruct-2507 (Prompt)	94
13	Amazon-Nova-Pro-v1:0 (FC)	93.5
13	Claude-Opus-4-5-20251101 (FC)	93.5
13	Command A (FC)	93.5
13	Gemini-2.5-Flash-Lite (FC)	93.5
13	Qwen3-32B (FC)	93.5
13	Qwen3-32B (Prompt)	93.5
19	Claude-Opus-4-5-20251101 (Prompt)	93
19	Grok-4-1-fast-non-reasoning (FC)	93
19	ToolACE-2-8B (FC)	93
19	palmyra-x-004 (FC)	93
19	xLAM-2-32b-fc-r (FC)	93
24	Claude-Haiku-4-5-20251001 (FC)	92.5
24	Grok-4-1-fast-reasoning (FC)	92.5
24	Mistral-Small-2506 (Prompt)	92.5
24	Qwen3-4B-Instruct-2507 (FC)	92.5
28	Qwen3-8B (FC)	92
28	xLAM-2-70b-fc-r (FC)	92
30	Amazon-Nova-2-Lite-v1:0 (FC)	91.5
30	GLM-4.6 (FC thinking)	91.5
30	Qwen3-30B-A3B-Instruct-2507 (FC)	91.5
33	Arch-Agent-3B	91
33	GPT-4.1-mini-2025-04-14 (FC)	91
33	Gemini-3-Pro-Preview (FC)	91
33	Gemma-3-12b-it (Prompt)	91
33	Hammer2.1-7b (FC)	91
38	Open-Mistral-Nemo-2407 (Prompt)	90.5
39	Gemini-2.5-Flash-Lite (Prompt)	90
39	Llama-3.3-70B-Instruct (FC)	90
41	Command A Reasoning (FC)	89.5
41	DeepSeek-V3.2-Exp (Prompt + Thinking)	89.5
41	mistral-large-2411 (FC)	89.5
44	GPT-5.2-2025-12-11 (FC)	89
44	Gemma-3-27b-it (Prompt)	89
44	Grok-4-0709 (Prompt)	89
47	CoALM-70B	88.5
47	Falcon3-10B-Instruct (FC)	88.5
47	Granite-3.2-8B-Instruct (FC)	88.5
47	Grok-4-0709 (FC)	88.5