Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Function Calling on BFCL Parity
Loading...
85.9
Accuracy
OpenThinkerAgent-32B
32.34
46.245
60.15
74.055
Jun 23, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
OpenThinkerAgent-32B
Train Size=100K, Metho...
2026.06
85.9
SWE-Lego-Qwen3-32B
Train Size=18K, Method...
2026.06
81
OT-Agent-SFT-8B (10K)
Base Model=OpenThinker...
2026.06
79.4
DeepSWE-Preview
Train Size=4.5K, Metho...
2026.06
77.2
SWE-Lego-Qwen3-8B
Base Model=Qwen3-8B, T...
2026.06
75.3
SA-SWE-32B
Train Size=4.5K, Metho...
2026.06
74.8
OT-Agent-ColdSFT+RL-8B
Base Model=OpenThinker...
2026.06
71.5
Nemotron-Terminal-32B
Train Size=264K, Metho...
2026.06
69.1
SERA-32B
Train Size=25K, Method...
2026.06
69.1
Qwen3-32B
Train Size=N/A, Method...
2026.06
68.3
OT-Agent-SFT-8B (100K)
Base Model=OpenThinker...
2026.06
65.9
Nemotron-Terminal-8B
Train Size=264K, Train...
2026.06
59.9
Endless Terminals
Base Model=Qwen3-8B, T...
2026.06
48.8
Qwen3-8B
Base Model=N/A, Train...
2026.06
34.4
Feedback
Search any
task
Search any
task