List your company

Mathematical Reasoning on APEX 2025

16.7AccuracyTRICE-30B
-0.6683.8418.3512.859May 7, 2026

Evaluation results

MethodLinks
TRICE-30BTool Usage=true, Parameter Scale=~30B, Evaluation Protocol=unified2026.0516.7
TRICE-4BTool Usage=true, Parameter Scale=<10B, Evaluation Protocol=unified2026.0513.9
GLM-4.7-Flash w/ recipeTool Usage=true, Parameter Scale=~30B, Evaluation Protocol=unified2026.0513.9
GLM-4.7-FlashTool Usage=true, Parameter Scale=~30B, Evaluation Protocol=unified2026.0511.1
Nemotron-3-Nano-30B-A3BTool Usage=true, Parameter Scale=~30B, Evaluation Protocol=unified2026.0511.1
GPT-OSS-20BTool Usage=true, Parameter Scale=~30B, Evaluation Protocol=unified2026.058.3
TRICE-4BTool Usage=false, Parameter Scale=<10B, Evaluation Protocol=unified2026.055.6
Qwen3-4B-Thinking-2507Tool Usage=false, Parameter Scale=<10B, Evaluation Protocol=unified2026.052.8
Qwen3-30B-A3B-Instruct-2507Tool Usage=false, Parameter Scale=~30B, Evaluation Protocol=unified2026.052.8
Qwen3.5-4BTool Usage=false, Parameter Scale=<10B, Evaluation Protocol=unified2026.050
Qwen3.5-9BTool Usage=false, Parameter Scale=<10B, Evaluation Protocol=unified2026.050
Qwen3-30B-A3B-Thinking-2507Tool Usage=false, Parameter Scale=~30B, Evaluation Protocol=unified2026.050
Qwen3.5-35B-A3BTool Usage=false, Parameter Scale=~30B, Evaluation Protocol=unified2026.050
TRICE-30BTool Usage=false, Parameter Scale=~30B, Evaluation Protocol=unified2026.050