Microsoft: Phi 4 Mini Instruct
Intelligence
—
Awaiting review
Speed
#216 / 254
45.0
tokens / sec
Input
#169 / 546
$0.080
per 1M tokens
Output
#209 / 546
$0.350
per 1M tokens
Context
#287 / 546
128,000
tokens
Analysis Summary
At $0.080 input and $0.350 output per 1M tokens, it is among the most expensive on the market. It offers a standard large context window.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Phi-4-mini-instruct is a lightweight open model built upon synthetic data and filtered publicly available websites - with a focus on high-quality, reasoning dense data. The model belongs to the Phi-4..
Performance Indices
Source: Artificial Analysis
8.4
Intelligence Index
3.6
Coding Index
8.2
Agentic Index
6.7
Math Index
Benchmark Scores
Intelligence
GPQA Diamond
33.1%
Graduate-level scientific reasoning
HLE
4.2%
Humanity's Last Exam
MMLU Pro
46.5%
Multi-task language understanding
MATH 500
69.6%
Mathematical problem-solving
AIME
3%
Competition mathematics
AIME 2025
6.7%
Competition mathematics (2025)
SciCode
10.8%
Scientific computing
Technical
LiveCodeBench
12.6%
Live coding evaluation
τ²-Bench
8.2%
Conversational agent benchmark
Content
IFBench
21.1%
Instruction following
LCR
13.7%
Long-context reasoning
Benchmark data from Artificial Analysis and Hugging Face
How does Microsoft: Phi 4 Mini Instruct stack up?
Compare side-by-side with other similar models.
Model Information
| OpenRouter ID |
microsoft/phi-4-mini-instruct
|
| Provider | microsoft |
| Release Date | October 17, 2025 |
| Context Length | 128,000 tokens |
| Max Completion | 128,000 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.08 | $0.000080 |
| Output | $0.35 | $0.000350 |
Live Performance
Live endpoint metrics — refreshed every 30 minutes.
340ms
Best Latency (TTFT)
156 tok/s
Best Throughput
0/1
Active Endpoints
Available via:
WandB
