Microsoft
Microsoft Research Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion...
Running this yourself: desktop gpu should be enough.
OpenRouter
Price record: 2026-11-18. Source: openrouter.
No current subscription pricing is tracked for this model.
Confirm this specific model, usage limits, and billing terms with the provider. A subscription does not automatically include API credits.
No login needed to compare. Prices are in USD; provider charges are separate from AI Market Cap plans. Context length, caching, tools, taxes, and regional terms can change the final cost. Open weights do not mean free hosting.
58.2
Quality Score
1221
Arena ELO
15B
Parameters
16K
Context
This measures the amount of verifiable public evidence we have, not how capable the model is. A missing field means it has not been verified yet, not that its value is zero.
19 of 22 public signals
Sign in to join the discussion
199.5K
Downloads
2.3K
Likes
Jan 2025
Released
5/5 signals
4/4 signals
5/5 signals
2/4 signals
3/4 signals
Parameters
14B
Training compute
9.3e23 FLOP
Dataset scale
Not reported
Base model
Not reported
Source-reported access: Open weights (unrestricted) · Confident confidence
Gaps we are still tracking
Launches
4
Benchmarks
20
Open Source
2
General
4
Recent launch, pricing, benchmark, and API signals linked to this model or its provider.
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
More ways to discover, access, and build with MAI models. MAI models are now available through Vercel, giving developers another way to bring Microsoft AI models into the products and experiences they’re building. This includes our existing MAI models plus today’s newest
View sourcelanguage: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
View source

MAI-Transcribe-2, the best transcription model in the world, is now generally available in Microsoft Foundry for production use cases! And it’s currently priced at just $0.10 per hour of audio. Come build with it!

More ways to discover, access, and build with MAI models. MAI models are now available through Vercel, giving developers another way to bring Microsoft AI models into the products and experiences they’re building. This includes our existing MAI models plus today’s newest
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Avg: 40.7 | IFEval: 68.8 | BBH: 55.3 | MATH: 50.0 | GPQA: 11.5 | MMLU-PRO: 48.6
Phi 4 is now available through local Ollama runtime. 128K context window listed. Phi-4-mini brings significant enhancements in multilingual support, reasoning, and mathematics, and now, the long-awaited function calling feature is finally supported.
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4
language: 0.2 | coding: 0.0 | instruction_following: 1.0 | Overall: 0.4