Skip to main content
AI Choice Engine

Mistral

Mistral Small 4

Mistral Small 4 is a Mistral model with 262K context, $0.15 in / $0.60 out per million tokens, last verified October 25, 2026. Open weights: yes. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.

Mistral Small 4 is Mistral's low-cost Apache-2.0 MoE — 6.5B active parameters with a 256K context and image input.

Recorded profile verification October 25, 2026Individual provider fields may change

AI model specification details
SpecificationMistral Small 4
ProviderMistral
TierBudget
Context window262K
Max outputNot verifiedUnverified
Input / 1M tokens$0.15
Output / 1M tokens$0.60
WeightsOpen
Parameters119B total / 6.5B active (MoE)
Reasoning levelsnone, high
Modalitiestext, image
LicenseApache 2.0
API model idmistral-small-2603
ReleasedMarch 16, 2026

Pricing tiers: Mistral API $0.15/$0.60 per MTok; cached input $0.015. Publisher checkpoint card declares Apache 2.0; 119B total / 6.5B active MoE, text/image input and text output. These are the default Standard USD values in the retained pricing view. Batch/Priority and regional controls are offered, but their exact tariffs and eligibility are not established by that view; do not assume a universal discount or residence guarantee.

Publisher evidence

Published benchmark results

Each result keeps its source date visible. Dates refer to source publication or capture unless the source explicitly states when it measured the result. A missing benchmark is not treated as a zero.

Published benchmark results for Mistral Small 4
BenchmarkScoreSource dateSource
Artificial Analysis Intelligence IndexMistral Small 4112026-10-01Artificial Analysis · View source

Best for

  • Cheap EU-hosted inference
  • Open-weight self-hosting
  • High-volume extraction

Watch out

None selects direct responses; high enables reasoning, and benchmark outcomes depend on this setting. A separate hard output cap is not established by the checked card. Residency and serving capabilities depend on the selected endpoint; parameter count alone does not establish local memory or throughput. The native card recommends temperature 0.7 for high reasoning and 0–0.7 for none; these are recommended settings, not memory or throughput guarantees.

History

Mistral Small 4 timeline

Release and status events for this model that passed two-source verification — dated, sourced, and linked.

Common questions

Mistral Small 4

Answered from the verified figures on this page rather than general guidance.

How much does Mistral Small 4 cost per million tokens?
Mistral Small 4 lists base rates of $0.15 per million input tokens and $0.60 per million output tokens. Estimate both input and output token volume and any additional charges for your workload. Mistral API $0.15/$0.60 per MTok; cached input $0.015. Publisher checkpoint card declares Apache 2.0; 119B total / 6.5B active MoE, text/image input and text output. These are the default Standard USD values in the retained pricing view. Batch/Priority and regional controls are offered, but their exact tariffs and eligibility are not established by that view; do not assume a universal discount or residence guarantee.
What is Mistral Small 4's context window?
Mistral Small 4 accepts about 262K tokens of context. That only matters if you routinely send very long documents, large codebases, or multi-turn histories that approach that limit.
What is Mistral Small 4 best for?
Mistral Small 4 is a budget tier from Mistral. It suits cheap eu-hosted inference, open-weight self-hosting, high-volume extraction. None selects direct responses; high enables reasoning, and benchmark outcomes depend on this setting. A separate hard output cap is not established by the checked card. Residency and serving capabilities depend on the selected endpoint; parameter count alone does not establish local memory or throughput. The native card recommends temperature 0.7 for high reasoning and 0–0.7 for none; these are recommended settings, not memory or throughput guarantees.
Can I self-host Mistral Small 4?
Mistral Small 4 publishes open weights, but self-hosting depends on the licence, hardware footprint, quantisation quality, and serving stack. A hosted API is often cheaper until you have measured throughput and concurrency on your own hardware.