Devstral 2

Frontier code agents model for software engineering tasks.

devstral-2512
STABLEGet StartedView uptime
262,144 context
Released December 9, 2025
Starting at $0.38/M (5% off) input tokens
Starting at $1.90/M (5% off) output tokens
Streaming
JSON Output
No ratings yetSign in to rate
5% offthis model via Mistral AI

Select Provider

All Providers for Devstral 2

PassingRight routes requests to the best providers that are able to handle your prompt size and parameters.

Context: 262.1k5% off
Input
$0.4$0.38
/M tokens
Cache Read
/M tokens
Output
$2$1.9
/M tokens
Get Started

Frequently asked questions

What is Devstral 2?

Frontier code agents model for software engineering tasks. You can access it through PassingRight's OpenAI-compatible API with automatic provider routing, fallback, and cost analytics.

How much does Devstral 2 cost?

Pricing for Devstral 2 on PassingRight starts at $0.40 per million input tokens and $2.00 per million output tokens, depending on the provider. The pricing table above always reflects the current per-provider rates.

What is the context length of Devstral 2?

Devstral 2 supports a context window of up to 262,144 tokens on its largest provider deployment.

Which providers serve Devstral 2?

Devstral 2 is served by Mistral AI through PassingRight. Requests are automatically routed to the best available provider, with fallback when a provider has issues.

Does Devstral 2 support tool calling and structured outputs?

Devstral 2 supports soft JSON output, but not tool calling or strict JSON output schema.

When was Devstral 2 released?

Devstral 2 was released on December 9, 2025.