Microsoft, open weights
Phi-4 Mini
Phi-4 Mini by Microsoft ranks 283rd of 354 ranked models on the Noometry Index as of October 2026, with a score of 30.9. Its strongest category is reasoning, where it ranks 195th. API pricing starts at $0.075 per million input tokens and $0.30 per million output tokens, with a 128K-token context window.
Last verified
Specifications
- Noometry rank
- #283 of 354
- Index score
- 30.9
- Evidence
- Reported 3 results
- Provider
Microsoft- Released
- December 11, 2024
- Weights
- Open weights
- Reasoning
- No
- Context window
- 128K
- Max output
- 4K
- Input price
- $0.075 / M
- Output price
- $0.30 / M
- Blended price
- $0.13 / M
- Output speed
- Not measured
- Value
- #35 of 219
- Knowledge cutoff
- October 2023
- Input
- text
Category scores
Each category score combines every public result we have in that category.
- Coding 28.1
- Reasoning 22.4
- Knowledge 25.3
Strengths and weaknesses
Categories where Phi-4 Mini places highest and lowest among the models ranked in each, with its score against that category's median.
Closest competitors
The models ranked just above and below Phi-4 Mini. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| Phi-4 | #279 | 31.2 | $0.0875 | — | Compare |
| Mistral Small 3.2 | #280 | 31.2 | $0.13 | 68 | Compare |
| Amazon Nova Pro | #281 | 31.0 | $1.40 | — | Compare |
| Llama 4 Maverick | #282 | 30.9 | $0.30 | 456 | Compare |
| Gemma 3 27B | #284 | 30.8 | $0.10 | 62 | Compare |
| Qwen1.5-72B | #285 | 30.8 | — | — | Compare |
| Granite 3.0 2b Instruct | #286 | 30.8 | — | — | Compare |
| Codellama 34b Instruct | #287 | 30.8 | — | — | Compare |
Sponsored placements are available on pages like this one. Advertise on Noometry
Benchmark results
Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.
Coding
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| SciCode | 10.8% | #121 of 121, top 100% | Epoch AI |
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| CritPt | 0% | #129 of 134, top 97% | Epoch AI |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| Vectara Hallucination Rate (lower is better) | 23.5% | #96 of 96, top 100% | Vectara Hallucination Leaderboard |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| azure | $0.075 | $0.30 | — | 2026-10-10 |
Compare Phi-4 Mini
- Phi-4 Mini vs Llama 4 Maverick
- Phi-4 Mini vs Gemma 3 27B
- Phi-4 Mini vs Amazon Nova Pro
- Phi-4 Mini vs Qwen1.5-72B
- Phi-4 Mini vs Mistral Small 3.2
- Phi-4 Mini vs Granite 3.0 2b Instruct
- Phi-4 Mini vs GPT-6 Astra
- Phi-4 Mini vs Claude Fable 5.1
- Phi-4 Mini vs Gemini 3.8 Flash
- Phi-4 Mini vs Kimi K3
- Phi-4 Mini vs Grok 4.6
- Phi-4 Mini vs Qwen3.8 Max
- Phi-4 Mini vs GLM-5.3
- Phi-4 Mini vs Muse Spark 1.3
Other Microsoft models
Frequently asked questions
How good is Phi-4 Mini?
Phi-4 Mini by Microsoft ranks 283rd of 354 ranked models on the Noometry Index as of October 2026, with a score of 30.9. Its strongest category is reasoning, where it ranks 195th. API pricing starts at $0.075 per million input tokens and $0.30 per million output tokens, with a 128K-token context window.
How much does Phi-4 Mini cost?
Phi-4 Mini costs $0.075 per million input tokens and $0.30 per million output tokens on azure.
What is Phi-4 Mini's context window?
Phi-4 Mini accepts up to 128K tokens of input and can write up to 4K tokens in one response.
Is Phi-4 Mini open source?
Yes. Phi-4 Mini's weights are downloadable; check the license for commercial terms.
What are Phi-4 Mini's strengths and weaknesses?
Relative to other ranked models, Phi-4 Mini places best in reasoning and lowest in coding.
What is Phi-4 Mini best at?
Its best category is reasoning, where it ranks 195th on Noometry.