List your company

Mathematical Reasoning on OlympBench

74.04Pass@1Qwen3-235B-A22B
21.166434.893248.6262.3468Dec 30, 2025Jan 22, 2026Feb 15, 2026Mar 10, 2026Apr 3, 2026Apr 26, 2026May 20, 2026

Evaluation results

MethodLinks
Qwen3-235B-A22BModel Category=Large Language Models, Thinking Mode=Thinking2025.1274.04
FIGRModel Category=Large Vision-Language Models2025.1272.4
Qwen3-32BModel Category=Large Language Models, Thinking Mode=Thinking2025.1271.96
Text-only RLModel Category=Large Vision-Language Models, Training=Reinforcement Learning (GRPO)2025.1270.33
GLM-4.5VModel Category=Large Vision-Language Models, Parameters=108B2025.1270.03
Qwen3-VL-32B-InstructModel Category=Large Vision-Language Models, Parameters=32B2025.1269.73
Qwen3-VL-8B-InstructModel Category=Large Vision-Language Models, Parameters=8B2025.1267.21
Qwen3-32BModel Category=Large Language Models, Thinking Mode=Non-Thinking2025.1252.08
RELEXBackbone=Qwen3-8B-Base, Cost=20%2026.0549.6
RLVRBackbone=Qwen3-8B-Base, Cost=100%2026.0549.5
Weight Extrap.Backbone=Qwen3-8B-Base, Cost=20%2026.0548.5
RELEXBackbone=Qwen3-4B-Base, Cost=15%2026.0547.4
ExPOBackbone=Qwen3-8B-Base, Cost=20%2026.0547.1
AlphaRLBackbone=Qwen3-8B-Base, Cost=20%2026.0546.9
RLVRBackbone=Qwen3-4B-Base, Cost=100%2026.0546.6
Logits Extrap.Backbone=Qwen3-8B-Base, Cost=20%2026.0546
ExPOBackbone=Qwen3-4B-Base, Cost=15%2026.0544.1
Weight Extrap.Backbone=Qwen3-4B-Base, Cost=15%2026.0542.9
AlphaRLBackbone=Qwen3-4B-Base, Cost=15%2026.0542
Logits Extrap.Backbone=Qwen3-4B-Base, Cost=15%2026.0540.1
BaseBackbone=Qwen3-8B-Base, Cost=0%2026.0536.9
BaseBackbone=Qwen3-4B-Base, Cost=0%2026.0531.6
RLVRBackbone=Qwen2.5-Math-1.5B, Cost=100%2026.0531.5
RELEXBackbone=Qwen2.5-Math-1.5B, Cost=15%2026.0531.4
Weight Extrap.Backbone=Qwen2.5-Math-1.5B, Cost=15%2026.0530.6
ExPOBackbone=Qwen2.5-Math-1.5B, Cost=15%2026.0529.5
AlphaRLBackbone=Qwen2.5-Math-1.5B, Cost=15%2026.0528.4
Logits Extrap.Backbone=Qwen2.5-Math-1.5B, Cost=15%2026.0528.2
BaseBackbone=Qwen2.5-Math-1.5B, Cost=0%2026.0523.2