TAU2
Measured May 14, 2026Source
Score
0.22
Models15 1 
Qwen3 Next 80B A3B Instruct
阿里巴巴阿里巴巴China
Qwen3 Next 80B A3B Instruct is a large language model from Alibaba's Qwen series, featuring a sparse activation architecture (likely 80B total parameters with ~3B active parameters per token). This design aims to deliver strong reasoning and coding capabilities while significantly improving inference speed and cost-efficiency compared to dense models of similar total size.
CodingReasoningFastCheapLong context
Input / 1M tokens
$0.50Output / 1M tokens
$2.00Output tokens/s
167.41First-token seconds
1.06sSupported plans
4Benchmark history
Evaluations
Terminalbench Hard
Measured May 14, 2026Source
Score
0.08
Lcr
Measured May 14, 2026Source
Score
0.51
Ifbench
Measured May 14, 2026Source
Score
0.4
Aime 25
Measured May 14, 2026Source
Score
0.66
Scicode
Measured May 14, 2026Source
Score
0.31
Livecodebench
Measured May 14, 2026Source
Score
0.68
Hle
Measured May 14, 2026Source
Score
0.07
Gpqa
Measured May 14, 2026Source
Score
0.74
Mmlu Pro
Measured May 14, 2026Source
Score
0.82
Artificial Analysis Math Index
Measured May 14, 2026Source
Score
66.3
Artificial Analysis Coding Index
Measured May 14, 2026Source
Score
15.3
Artificial Analysis Intelligence Index
Measured May 14, 2026Source
Score
20.1
Aime
Measured May 14, 2026Source
Score
0.94
Math 500
Measured May 14, 2026Source
Score
0.98
Plan availability
Products and plans that support this model
io.net Plan
io.net Plan
io.net is a decentralized GPU cloud platform providing subscription plans and pay-as-you-go access to a variety of AI models for chat, coding, content creation, and automation workflows.
Discussion
Thinking... Make sure you are connected to GitHub server