TAU2
Measured May 14, 2026Source
Score
0
Models13 4 0 
DeepSeek V3.2 Speciale
DeepSeekDeepSeekChina
A specialized variant of the DeepSeek V3.2 architecture, optimized for enhanced performance in code generation and complex reasoning tasks. It maintains the series' hallmark of high cost-efficiency and rapid response times.
CodingReasoningFastCheap
Input / 1M tokens
$0.00Output / 1M tokens
$0.00Supported plans
0Benchmark history
Evaluations
Terminalbench Hard
Measured May 14, 2026Source
Score
0.35
Lcr
Measured May 14, 2026Source
Score
0.59
Ifbench
Measured May 14, 2026Source
Score
0.64
Aime 25
Measured May 14, 2026Source
Score
0.97
Scicode
Measured May 14, 2026Source
Score
0.44
Livecodebench
Measured May 14, 2026Source
Score
0.9
Hle
Measured May 14, 2026Source
Score
0.26
Gpqa
Measured May 14, 2026Source
Score
0.87
Mmlu Pro
Measured May 14, 2026Source
Score
0.86
Artificial Analysis Math Index
Measured May 14, 2026Source
Score
96.7
Artificial Analysis Coding Index
Measured May 14, 2026Source
Score
37.9
Artificial Analysis Intelligence Index
Measured May 14, 2026Source
Score
29.4
CODPL speed
Provider ranking
百度云(千帆)
Product: 百度千帆 Coding Plan32.15
1,030 ms
100%
3
1h
Rank #4
41.26
962 ms
100%
3
1h
Rank #2
Ollama Cloud
Product: Ollama Cloud45.43
1,501 ms
100%
2
1h
Rank #3
61.91
733 ms
100%
3
1h
Rank #1
Plan availability
Products and plans that support this model
No products or plans have been linked to this model yet.
Discussion
Thinking... Make sure you are connected to GitHub server