List your company

Scientific Reasoning on GPQA Diamond (avg@3)

82.2avg@3Nanbeige4-3B-Thinking
61.19266.64672.177.554Dec 6, 2025

Evaluation results

MethodLinks
Nanbeige4-3B-ThinkingRelease Identifier=25112025.1282.2
Qwen3-30A3-2507Release Identifier=25072025.1273.4
Qwen3-32B-2504Release Identifier=25042025.1268.7
Qwen3-4B-2507Release Identifier=25072025.1267.2
Qwen3-14B-2504Release Identifier=25042025.1264
Qwen3-8B-2504Release Identifier=25042025.1262