Skip to main content

    Microsoft

    Phi-4-reasoning

    Microsoft's 14B dense reasoning model, released April 30, 2025 under the MIT license. It accepts and produces text with a 32K-token context. Supervised fine-tuning emphasizes mathematical reasoning, science and coding. Designed for reasoning research and applications with constrained memory or compute, with downloadable weights for local deployment.

    32K tokens
    Text only
    Pro
    Provider
    Microsoft
    Context
    32K
    Estimated · As of Aug 11, 2026Stale · 31d old
    yno subscription tier
    Pro
    Released
    Apr 30, 2025
    Speed
    Fast
    Reasoning
    Expert
    Modality
    Text only

    Benchmarks

    Quality

    AA Intelligence Index
    16points
    Estimated · As of Aug 11, 2026Stale · 31d old
    GPQA Diamond
    80%
    Estimated · As of Aug 11, 2026Stale · 31d old

    Speed and latency

    Output speed
    142tokens/s
    Estimated · As of Aug 11, 2026Stale · 31d old

    API pricing

    Not available

    AA measurements unavailable

    Indexes use points; evaluations use accuracy percentages. API measurements do not measure yno application performance. Methodology

    Capabilities

    • Reasoning
    • Small Model
    • Open Source
    • Math
    • Code
    • Fine-tuning

    Use cases

    • Edge reasoning
    • Cost-sensitive inference
    • Fine-tuned deployments
    • Local agents

    Strengths

    • 14B dense parameters
    • 32K context
    • MIT-licensed weights
    • Reasoning-focused fine-tuning

    Best for

    Teams evaluating a locally deployable 14B model for mathematical reasoning

    Use Phi-4-reasoning in yno.ai

    No credit card required

    Related models

    Frequently asked

    What is Phi-4-reasoning?
    Microsoft's 14B dense reasoning model, released April 30, 2025 under the MIT license. It accepts and produces text with a 32K-token context. Supervised fine-tuning emphasizes mathematical reasoning, science and coding. Designed for reasoning research and applications with constrained memory or compute, with downloadable weights for local deployment.
    How much does Phi-4-reasoning cost in yno.ai?
    Phi-4-reasoning is available on the Pro tier of yno.ai.
    What can Phi-4-reasoning do?
    Phi-4-reasoning is best for Teams evaluating a locally deployable 14B model for mathematical reasoning. Its main capabilities include Reasoning, Small Model, Open Source, Math, Code, Fine-tuning.
    How does Phi-4-reasoning compare to other models?
    Phi-4-reasoning excels at 14B dense parameters and is recommended for Edge reasoning, Cost-sensitive inference, Fine-tuned deployments, Local agents. See related models below.
    How do I use Phi-4-reasoning in yno.ai?
    Sign up for yno.ai, select Phi-4-reasoning from the model picker, and start chatting.