TAU2
Measured Jul 10, 2026Source
Score
0.22
Models17 2 
DeepSeek R1 Distill Llama 70B
DeepSeekDeepSeekChina
A distilled version of the DeepSeek R1 reasoning model, built on the Llama 70B architecture. It inherits strong reasoning and chain-of-thought capabilities from the R1 series while being optimized for efficiency. The model excels at complex problem-solving, code generation, and tasks requiring logical deduction.
CodingReasoningFastCheap
Input / 1M tokens
$0.70Output / 1M tokens
$1.05Output tokens/s
29.88First-token seconds
0.5sSupported plans
4Benchmark history
Evaluations
Terminalbench Hard
Measured Jul 10, 2026Source
Score
0.02
Lcr
Measured Jul 10, 2026Source
Score
0.11
Ifbench
Measured Jul 10, 2026Source
Score
0.28
Aime 25
Measured Jul 10, 2026Source
Score
0.54
Aime
Measured Jul 10, 2026Source
Score
0.67
Math 500
Measured Jul 10, 2026Source
Score
0.93
Scicode
Measured Jul 10, 2026Source
Score
0.31
Livecodebench
Measured Jul 10, 2026Source
Score
0.27
Hle
Measured Jul 10, 2026Source
Score
0.06
Gpqa
Measured Jul 10, 2026Source
Score
0.4
Mmlu Pro
Measured Jul 10, 2026Source
Score
0.8
Artificial Analysis Math Index
Measured Jul 10, 2026Source
Score
53.7
Artificial Analysis Intelligence Index
Measured Jul 10, 2026Source
Score
9.9
Tau Banking
Measured Jul 10, 2026Source
Score
0.06
Terminalbench V2 1
Measured Jul 10, 2026Source
Score
0.19
Artificial Analysis Coding Index
Measured Jul 10, 2026Source
Score
24.6
Plan availability
Products and plans that support this model
联通云大模型 Coding Plan
联通云大模型 Coding Plan
China Unicom Cloud's Coding Plan is a subscription service for AI-powered coding, providing access to multiple large language models via an API for use with mainstream AI programming tools. It offers tiered plans with monthly request quotas.
联通元景 Token Plan
联通元景 Token Plan
China Unicom Cloud's Unicom Yuanjing Token Plan is a subscription service providing token-based access to multiple AI models, including DeepSeek and QwQ, designed for AI coding and development scenarios. It offers flexible monthly packages with varying token quotas for individuals and teams.
User ratings
Loading ratings...
Discussion
Thinking... Make sure you are connected to GitHub server