Text & chat

    FRFranceMistral Large 4 (Le Chonk) — Preview

    Mistral AI · France · EU / EEA

    Teams evaluating a European multimodal model via API for document analysis, coding and agentic workflows. Mistral documents a 1M-token context window; Artificial Analysis lists 524k for the evaluated preview.

    EU company
    EU data residency
    Compliance: High
    Open Mistral Large 4 (Le Chonk) — Preview

    Classification

    Jurisdiction tier
    EU / EEA — EU member state or EEA (Norway, Iceland, Liechtenstein) — GDPR applies directly.
    Note
    Mistral states that the public preview is served from its own European datacenters. Verify the data-processing terms for your API deployment.
    Caveat
    Public API preview, launched 6 October 2026 — not a ready-made chat product. Model weights are promised by the end of October but are not yet available at this review; self-hosting and the eventual weights licence must be verified after release. Not labelled open-weight or self-hostable.
    Compliance level
    High — Derived from the verified classification data on this site: Very high = EU/EEA jurisdiction + documented EU data residency + self-hostable option. High = EU/EEA + EU data residency. Medium = adequacy country + residency, or EU/EEA without documented residency. Each underlying claim carries its own source.
    Tags
    enterprise
    multimodal
    api
    preview
    What the badges mean

    EU company — Vendor headquartered in Europe (state country + tier).

    EU data residency — EU data-residency / processing option available.

    Why these rules
    Last verified
    2026-10-09

    Sources

    AI Capability Snapshot

    AI Capability Index — normalised so the strongest model scores 100 (anchor: Claude Fable 5.1 (max effort)). How this is measured

    Capability index58 / 100
    Coding subscoreNot yet scored

    DeepSWE, the Coding Agent Index and cybersecurity benchmarks are different evaluations, not substitutes for SWE-bench Verified or the Intelligence Index. No coding score is inferred from them. The preview may change before the weights release.

    Scored on
    Mistral Large 4 Preview. Artificial Analysis Intelligence Index 38 → 58/100, rounded against the existing AA 66 anchor. No verified SWE-bench Verified result was found at this review, so the coding subscore remains empty.
    Retrieved
    2026-10-09

    Reference models (non-EU)

    For scale only — these models are not European and are not listed in the directory.

    • Claude Fable 5.1 · Anthropic (US)100 / 100
    • Claude Opus 5 · Anthropic (US)95 / 100
    • GPT-5.6 Sol · OpenAI (US)92 / 100
    • Gemini 3.7 Flash · Google (US)85 / 100

    Independent, informational guide — not legal, security or compliance advice. Classifications describe company location, GDPR transfer status, data-residency options and licensing as of the date shown, from public sources. Verify data-processing terms with the provider before use, especially for personal or sensitive data. No affiliation with any vendor.

    Back to the directoryFlag as outdated