Share your thoughts, 1 month free Claude Pro on usSee more

General Knowledge Reasoning on MMLU Pro (Accuracy)

82.93Accuracy

Qwen3-Next-80B-A3B-Instruct

Updated 22d ago

Evaluation Results

Method	Links
Qwen3-Next-80B-A3B-Instruct 2026.03		82.93
ParaManager-SFT 2026.04		81.43
ParaManager 2026.04		81.43
Puppeteer 2026.04		80.54
Qwen3-Coder 30B-A3B-Instruct 2026.04		80.36
GPT-OSS-20B 2026.04		80
GPT-OSS-20B 2026.04		80
GPT-OSS-20B 2026.04		80
Qwen3-Coder 30B-A3B-Instruct 2026.04		80
Router 2026.04		80
ParaManager-Serial 2026.04		80
Qwen3-Omni-A3B-Instruct 2026.03		79.89
EvoFlow 2026.04		79.82
GPT-OSS-20B 2026.04		79.64
ParaManager-Mono 2026.04		79.64
Meta Agent Search 2026.04		79.46
ToolOrchestra 2026.04		79.46
Qwen3-30B A3B-Instruct-2507 2026.04		78.57
Nemotron 3 Nano 30B-A3B 2026.04		78.3
Qwen3-30B A3B-Instruct-2507 2026.04		77.86
Nemotron 3 Nano Omni 2026.04		77.3
Qwen3-Coder 30B-A3B-Instruct 2026.04		77.14
LongCat-Next 2026.03		77.02
Qwen3-30B A3B-Instruct-2507 2026.04		76.96
Qwen3-30B A3B-Instruct-2507 2026.04		76.79
Qwen3-4B Instruct-2507 2026.04		75.71
Qwen3-Coder 30B-A3B-Instruct 2026.04		75.71
Qwen3-4B Instruct-2507 2026.04		74.82
Qwen3-4B Instruct-2507 2026.04		74.11
Qwen3-4B Instruct-2507 2026.04		72.86
Kimi-Linear-48B-A3B 2026.03		67.22
HeRL 2026.03		62.1
Qwen3-Omni 2026.04		61.6
Qwen3-4B-Instruct-2507 2026.03		61.5
INTUITOR 2025.05		61.3
Qwen3-14B 2025.05		59.7
GRPO 2026.05		59.05
INTUITOR 2025.05		58.3
SPIRAL 2026.05		58.28
STRATAGEM 2026.04		57.83
GRPO 2025.05		57.8
MARS 2026.05		57.8
DEPT 2026.05		57.6
Qwen2.5-7B-Instruct 2026.03		57.3
Qwen2.5-14B 2025.05		56.5
DEPT 2026.05		56.45
HeRL 2026.03		55.5
SPIRAL 2026.04		53.93
MARS 2026.05		53.34
SPAG 2026.05		52.71
SPAG 2026.05		51.46
INTUITOR 2025.05		51.4
SPIRAL 2026.05		51.32
GRPO 2025.05		51.1
PSFT 2025.08		50.55
Qwen2.5-7B 2025.05		49.7
GRPO 2026.05		49.14
Base 2025.08		47.9
Qwen3-4B-Base 2026.04		47.2
VANILLA 2026.05		46.97
SFT 2025.08		40.67
VANILLA 2026.05		39.36
HeRL 2026.03		34.8
Llama-3.2-3B-Instruct 2026.03		33.9