live
weekly refresh
basedagi.org
benchmark evidence

BFCL Non-Live

BFCL non-live function calling accuracy.

winner on BFCL Non-Live
direct benchmark result, not a broad vertical composite | source row dated 2026-05-15
scored on 2026-05-15
latest mapped results | top 20
#ModelScoreEvidenceTested
1Mistral: Mistral Small 4
Mistralai
89.7
model-only
independent_benchmark
2026-05-15
2Anthropic: Claude Opus 4.5
Anthropic
89.7
model-only
independent_benchmark
2026-05-15
3Meta: Llama 4 Scout
Meta Llama
89.4
model-only
independent_benchmark
2026-05-15
4OpenAI: GPT-4.1
Openai
88.7
model-only
independent_benchmark
2026-05-15
5Meta: Llama 4 Maverick
Meta Llama
88.7
model-only
independent_benchmark
2026-05-15
6Google: Gemini 2.5 Flash
Google
88.1
model-only
independent_benchmark
2026-05-15
7Meta: Llama 3.3 70B Instruct
Meta Llama
88.0
model-only
independent_benchmark
2026-05-15
8Google: Gemma 3 27B
Google
87.2
model-only
independent_benchmark
2026-05-15
9Amazon: Nova Pro 1.0
Amazon
86.6
model-only
independent_benchmark
2026-05-15
10Mistral Large
Mistralai
83.0
model-only
independent_benchmark
2026-05-15
11MoonshotAI: Kimi K2 Thinking
Moonshotai
81.6
model-only
independent_benchmark
2026-05-15
12OpenAI: o4 Mini
Openai
81.3
model-only
independent_benchmark
2026-05-15
13OpenAI: GPT-5.2
Openai
78.3
model-only
independent_benchmark
2026-05-15
14Anthropic: Claude Sonnet 4.5
Anthropic
59.8
model-only
independent_benchmark
2026-05-15
15Anthropic: Claude Haiku 4.5
Anthropic
55.4
model-only
independent_benchmark
2026-05-15
16OpenAI: o3
Openai
40.4
model-only
independent_benchmark
2026-05-15
17DeepSeek: DeepSeek V3.2
Deepseek
34.9
model-only
independent_benchmark
2026-05-15
what this result means

BFCL non-live function calling accuracy.

This benchmark contributes direct public evidence. Read its scope before generalizing the result.

A win here is a win on BFCL Non-Live. Broad task pages require independent corroboration before naming a general winner.

source record
category: structured_output
metric: accuracy
matched models: 17
latest source date: 2026-05-15
direction: higher is better
inspect upstream source ->