List your company
ResearchFormerly WizWand

Mathematical Reasoning on Math Reasoning Suite Arithmetic Mean

78.1Average Score (@16)THINC-4B
23.60437.75251.966.048May 8, 2026
Updated 5mo ago

Evaluation results

MethodLinks
THINC-4BReasoning Type=THINC Student2026.0578.1
Qwen3-235B-A22B-ThinkingReasoning Type=NL-Only2026.0575.2
ASTER-4BReasoning Type=Tool-Integrated2026.0573.8
Qwen3-4B-Thinking-2507Reasoning Type=NL-Only2026.0565
gpt-oss-20B (medium)Reasoning Type=NL-Only2026.0564.7
Qwen3.5-27B*Reasoning Type=THINC Teacher, Python Interpreter=prompted2026.0564.7
OpenReasoning-Nemotron-7BReasoning Type=NL-Only2026.0562.7
Qwen3-4B-Thinking-2507*Reasoning Type=Tool-Integrated, Python Interpreter=prompted2026.0562.6
rStar2-Agent-14BReasoning Type=Tool-Integrated2026.0561.5
DemyAgent-4BReasoning Type=Tool-Integrated2026.0557.3
ReTool-32BReasoning Type=Tool-Integrated2026.0550.7
THINC-1.7BReasoning Type=THINC Student2026.0542.8
Qwen3-1.7BReasoning Type=NL-Only2026.0532.2
Qwen3-1.7B*Reasoning Type=Tool-Integrated, Python Interpreter=prompted2026.0529.8
CoRT-1.5BReasoning Type=Tool-Integrated2026.0525.7