Skip to main content

    Mistral AI

    Mistral Small 4

    Mistral's March 16, 2026 unified model: merges Magistral (reasoning), Pixtral (vision), and Devstral (agentic coding) into a single 119B-total MoE with 6.5B active per token. 256K context, Apache 2.0, configurable reasoning effort.

    256K tokens
    Multimodal
    Pro
    Provider
    Mistral AI
    Context
    256K
    Catalog · As of Aug 11, 2026Stale · 31d old
    yno subscription tier
    Pro
    Released
    Mar 16, 2026
    Speed
    Fast
    Reasoning
    Advanced
    Modality
    Multimodal

    Benchmarks

    Quality

    AA Intelligence Index
    11.5points
    Artificial Analysis · Date unknownAge unknown
    AA Coding Index
    26.6points
    Artificial Analysis · Date unknownAge unknown
    GPQA Diamond
    76.9%
    Artificial Analysis · Date unknownAge unknown

    Speed and latency

    Output speed
    169.4tokens/s
    Configuration: Prompt length: 1,000 · Parallel queries: 1
    Artificial Analysis · Date unknownAge unknown
    Time to first token
    0.55s
    Configuration: Prompt length: 1,000 · Parallel queries: 1
    Artificial Analysis · Date unknownAge unknown
    Time to first answer token
    12.35s
    Configuration: Prompt length: 1,000 · Parallel queries: 1
    Artificial Analysis · Date unknownAge unknown

    API pricing

    API input cost
    $0.15/ 1M tokens
    Artificial Analysis · Date unknownAge unknown
    API output cost
    $0.60/ 1M tokens
    Artificial Analysis · Date unknownAge unknown

    Benchmark variant: Mistral Small 4 (Reasoning)

    AA data retrieved Sep 11, 2026 · Artificial Analysis

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

    Capabilities

    • Reasoning
    • Vision
    • Agentic
    • Code
    • Configurable

    Use cases

    • All-in-one deployments
    • Agentic coding
    • Visual analysis
    • European sovereignty

    Strengths

    • Unified reasoning+vision+coding
    • Configurable reasoning depth
    • 6.5B active of 119B MoE
    • Apache 2.0

    Best for

    Teams that want one model for reasoning, vision, AND agentic coding without operating three

    Use Mistral Small 4 in yno.ai

    No credit card required

    Related models

    Frequently asked

    What is Mistral Small 4?
    Mistral's March 16, 2026 unified model: merges Magistral (reasoning), Pixtral (vision), and Devstral (agentic coding) into a single 119B-total MoE with 6.5B active per token. 256K context, Apache 2.0, configurable reasoning effort.
    How much does Mistral Small 4 cost in yno.ai?
    Mistral Small 4 is available on the Pro tier of yno.ai.
    What can Mistral Small 4 do?
    Mistral Small 4 is best for Teams that want one model for reasoning, vision, AND agentic coding without operating three. Its main capabilities include Reasoning, Vision, Agentic, Code, Configurable.
    How does Mistral Small 4 compare to other models?
    Mistral Small 4 excels at Unified reasoning+vision+coding and is recommended for All-in-one deployments, Agentic coding, Visual analysis, European sovereignty. See related models below.
    How do I use Mistral Small 4 in yno.ai?
    Sign up for yno.ai, select Mistral Small 4 from the model picker, and start chatting.