Mistral AI
Mistral Small 4
Mistral Small 4 is a model with open weights from Mistral AI. This model has not been measured on the Vals AI Legal Research Bench. The compliance gate has not yet been assessed for this model.
- Status unknown
- Open weights
- 16 March 2026
- PrudentScore
- -No score: the gate is not open
- Compliance gate
- Not yet assessedMistral EU endpoint (api.eu.mistral.ai; Mistral AI is a French SA in Paris): the documentation says regional endpoints only serve models hosted in that region, and whether this model is there cannot be measured without a Mistral key. Commercial Terms of Service (effective 2026-09-25, read 2026-11-10) section 4.2: Mistral does not train on Customer Data or Outputs except (a) on opt-in for a product that is opt-out by default or without opt-out on a product that is opt-in by default, (b) feedback, (c) as provided in an Order Form, (d) for Labs or Preview Models; whether this model or endpoint falls under an opt-in default has not been verified
- Cost per task
- -
01
Compliance gate
Why
- Mistral EU endpoint (api.eu.mistral.ai; Mistral AI is a French SA in Paris): the documentation says regional endpoints only serve models hosted in that region, and whether this model is there cannot be measured without a Mistral key. Commercial Terms of Service (effective 2026-09-25, read 2026-11-10) section 4.2: Mistral does not train on Customer Data or Outputs except (a) on opt-in for a product that is opt-out by default or without opt-out on a product that is opt-in by default, (b) feedback, (c) as provided in an Order Form, (d) for Labs or Preview Models; whether this model or endpoint falls under an opt-in default has not been verified.
- Open weights, but no route found at Vertex AI (EU), Azure Foundry (Data Zone EU), AWS Bedrock (EU regions) or the Mistral EU endpoint: the gate only opens on a dedicated EU endpoint or self-hosting, where the region is fixed by contract (as in the existing open-weights assessment).
The gate only opens when all three requirements are covered: processing in the EU, a contracting party in the EU or EEA, and a contractual exclusion of training on customer data. The gate tests a route, not a model name: the same model can pass through one provider and fail through another. Where a provider publishes no figure, we give no estimate.
Sources for this verdict
- docs.mistral.ai/inference/regional-inference
- legal.mistral.ai/terms/commercial-terms-of-service/
- docs.cloud.google.com/gemini-enterprise-agent-platform/resources/data-residency
- learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/models-sold-directly-by-azure-region-availability
- docs.aws.amazon.com/bedrock/latest/userguide/models-region-compatibility.html
- docs.mistral.ai/inference/regional-inference
02
Routes and EU processing
The gate tests a route (processor, region and contract), not a model name. Per route: processing in the EU, the contracting party and training on customer data.
| Processor | Region or deployment | EU processing | Source |
|---|---|---|---|
| Mistral AI (api.eu.mistral.ai) | EU endpoint exists (api.eu.mistral.ai, 1.1x surcharge); whether this model runs on the EU endpoint can only be measured with a Mistral key (models.list) | Unknown | SourceRead 10 November 2026 |
03
Scores per benchmark
Every measurement with its source, measurement date and the day we read it. The position counts only the models measured on the same benchmark and version.
- Legal Research BenchNot measured
- Harvey LAB (Vals implementation)Not measured
- LegalBench (Vals leaderboard)Not measured
- CaseLaw v2Not measured
- TaxEval v2Not measured
- Tax Agent BenchNot measured
- CorpFin v2Not measured
- Immigration BenchNot measured
- Access to JusticeNot measured
- Public Benefits BenchNot measured
- Vals Legal IndexNot measured
- Vals IndexNot measured
- GPQA Diamond (Vals-run)Not measured
- MMLU Pro (Vals-run)Not measured
- Artificial Analysis Intelligence IndexNot measured
The PrudentScore is a composition by Prudai B.V. based on publicly published figures and does not represent the view or endorsement of Vals AI, Harvey or Artificial Analysis. Figures from Artificial Analysis appear in running text only, with attribution and a link, as that party's terms require [9].
04
Key facts
- Developer
- Mistral AI
- Model family
- Mistral
- Release date
- 16 March 2026
- Status
- Status unknown
- Context window
- 262K tokens
- Max output tokens
- 256K tokens
- List price
- $0.15 input · $0.60 outputper million tokens · Read 10 November 2026
- Weights
- Open
- = not yet recorded in the register.
05
Frequently asked questions about this model
Does Mistral Small 4 pass the compliance gate for legal work?
That has not yet been assessed for this model.
What does Mistral Small 4 cost per task?
The list price is $0.15 per million input tokens and $0.60 per million output tokens.
Related models
- Mistral Small 3.1 (03/2025)
- Muse Spark
- Mistral Medium 3.523
- Mistral Large 4 (GA, 26.10)
- Mistral Large 4 (preview)
- Devstral 2
Claude and Anthropic, GPT and OpenAI, Gemini and Google, Grok and xAI, Kimi and Moonshot AI, Muse and Meta, Mistral, DeepSeek, GLM and Z.ai, Harvey, Vals AI and Artificial Analysis are trademarks of their respective owners. They are named for identification only and this implies no partnership, sponsorship or endorsement.
Brand marks appear for identification only and imply no partnership, sponsorship or endorsement.