Mistral Small 4

Mistral Small 4 is Mistral’s current Small-model reference: a European open-model lane entry for efficient general work, coding/reasoning hybrids, and official Mistral API cost comparisons.

What this model is

Mistral’s models overview describes Mistral Small 4 as an open model and a hybrid model that unifies instruct, reasoning, and coding in a single efficient model. The same overview points users from the deprecated Mistral Small 3.2 entry to Mistral Small 4 as the alternative.

Mistral Small 4 sits in the “small but capable” lane, where the model should be judged against Qwen, DeepSeek, GLM, Kimi, GPT-4o, and Claude Sonnet on cost, capability, and deployment preference.

Pricing notes

The calculator on this page uses Mistral’s official API rate for mistral-small-latest / Mistral Small 4: $0.15 per 1M input tokens and $0.60 per 1M output tokens.

This is a direct Mistral API price reference, not an OpenRouter or reseller route. If Mistral changes model aliases or pricing, recheck Mistral’s official pricing page before relying on these numbers for production planning.

Benchmarks and specs

Mistral’s models overview identifies Mistral Small 4 as an open model and describes it as a hybrid model for instruct, reasoning, and coding. This page does not list specific benchmark scores until they are tied to a named Mistral source and test setup.

No verified context-window value for Mistral Small 4 is attached to this page yet, so no context number is shown in the page header.

Best fit

Mistral Small 4 is best for cost-sensitive European API comparisons, small-model routing tests, coding/reasoning hybrid work, and alternatives to larger frontier reference models.

It is a useful “don’t overpay by default” comparison: try the smaller official Mistral route when the task does not need GPT-4o or Claude Sonnet-level spend.

Sources

• Mistral API pricing
• Mistral models overview