TAU2
Measured May 14, 2026Source
Score
0
Models15 0 
Phi-4
MicrosoftMicrosoftUnited States
Phi-4 is a small language model from Microsoft's Phi series, designed for strong reasoning and coding capabilities while maintaining low latency and cost. It is optimized for efficiency and practical deployment in resource-constrained environments.
CodingReasoningFastCheap
Input / 1M tokens
$0.125Output / 1M tokens
$0.50Output tokens/s
37.74First-token seconds
0.54sSupported plans
0Benchmark history
Evaluations
Terminalbench Hard
Measured May 14, 2026Source
Score
0.04
Lcr
Measured May 14, 2026Source
Score
0
Ifbench
Measured May 14, 2026Source
Score
0.24
Aime 25
Measured May 14, 2026Source
Score
0.18
Aime
Measured May 14, 2026Source
Score
0.14
Math 500
Measured May 14, 2026Source
Score
0.81
Scicode
Measured May 14, 2026Source
Score
0.26
Livecodebench
Measured May 14, 2026Source
Score
0.23
Hle
Measured May 14, 2026Source
Score
0.04
Gpqa
Measured May 14, 2026Source
Score
0.58
Mmlu Pro
Measured May 14, 2026Source
Score
0.71
Artificial Analysis Math Index
Measured May 14, 2026Source
Score
18
Artificial Analysis Coding Index
Measured May 14, 2026Source
Score
11.2
Artificial Analysis Intelligence Index
Measured May 14, 2026Source
Score
10.4
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