EN / 中文
Model prices / Mistral AI / Devstral 2

Devstral 2DeprecatedOpen weights

Mistral AI · released 2025-12-09

The official API price of Devstral 2 is $0.4 per 1M input tokens and $2 per 1M output tokens (as of 2026-11-12). Context window: 262K tokens, up to 262K output tokens per request. 14 providers sell it. The lowest is NanoGPT: $0.4 input, $1.4 output. This model is marked as deprecated.

Price by providerUSD per 1M tokens

ProviderInputOutputCachedNotes
MistralOfficialDeprecateddevstral-2512
$0.4$2—OfficialDeprecatedOfficial docs
NNanoGPTLowestmistralai/devstral-2-123b-instruct-2512
$0.4$1.4$0.2LowestOfficial docs
EEden AImistral/devstral-2512
$0.4$2$0.04Official docs
KKilo Gatewaymistralai/devstral-2512
$0.4$2$0.04Official docs
LLLM Gatewaymistral/devstral-2512
$0.4$2—Official docs
MMerge Gatewaymistral/devstral-2512
$0.4$2$0.04Official docs
OpenRoutermistralai/devstral-2512
$0.4$2$0.04Official docs
EEden AImistral/devstral-2512@eu
$0.44$2.2$0.044Official docs
AI/ML APImistralai/devstral-2512
$0.46552$2.3276$0.046552Official docs
CCortecsdevstral-2512
$0.478$2.392$0.045Official docs
Amazon BedrockDeprecatedmistral.devstral-2-123b
$0.4$2—DeprecatedOfficial docs
DevPass (LLM Gateway)Deprecateddevstral-2512
$0.4$2—DeprecatedOfficial docs
3 more listings
ProviderInputOutputCachedNotes
GreenPTDeprecateddevstral-2-123b-instruct-2512
$0.57$2.736—DeprecatedOfficial docs
IInferXDevstral-2-123B-Instruct-2512-int4-AutoRound
$0$0—Official docs
AAnyAPIDeprecatedmistralai/devstral-2512
———DeprecatedOfficial docs

Providers are those listed by models.dev. This site has not vetted them one by one; each provider's own page has the binding price.

About

Mistral's coding-agent model for repository work, terminal tasks, and software fixes

Description from models.dev.

Specs

Type
Text
Context
262K
Max output
262K
Input
Text
Output
Text
Reasoning
No
Tool calling
Yes
Structured output
Yes
Knowledge cutoff
2025-12
Open weights
Yes
Released
2025-12-09

Source

Prices and specs come from models.dev (MIT licence), updated 2026-11-12. Each provider's name links to its official docs. A provider priced at 0 usually includes the model in a subscription; it is not necessarily a free API.