Plain-English Summary
A compact reasoning model for efficient math, logic, and educational applications.
Phi-4-mini-reasoning
A compact reasoning model for efficient math, logic, and educational applications.
Phi-4
Reasoning LLM
GA / open weights
Open weights: Yes
API: Yes and self-hosted
Technical Notes
Small reasoning-tuned model with up to 128K input and output capacity, designed for efficient deployment.
Best For
Education, math helpers, edge reasoning, private applications
Watch Out For
Lower general capability and knowledge than larger reasoning models.
Model Specs
- Maker
- Microsoft
- Country
- USA
- Family
- Phi-4
- Type
- Reasoning LLM
- Release date
- 2025-04-30
- Date confidence
- Exact
- Status
- GA / open weights
- Context window
- 128,000 tokens
- Max output
- 128,000
- Parameters
- Not disclosed
- Active parameters
- Not disclosed
- Architecture
- Dense reasoning transformer
- Reasoning
- Yes — specialized
- Tool calling
- Host dependent
- Structured output
- Host dependent
- API available
- Yes and self-hosted
- Open weights
- Yes
- Self-hostable
- Yes
- License
- MIT
- Inputs
- Text
- Outputs
- Text
- Approx. price
- See provider pricing
- How to access
- Provider website
- Model / API ID
- Phi-4-mini-reasoning
- Knowledge cutoff
- Not publicly disclosed
- Fine-tuning
- Fine-tuning, quantization, self-hosting
- Completeness
- 90 %
- Verified
- Jul 18, 2026
- Snapshot date
- Jul 18, 2026
- Source
- Official source