modelgrep
M

Mistral: Devstral 2 2512

mistralai/devstral-2512

113th smartest of 178ToolsJSON
Use via OpenRouter ↗
Intelligence
22.0
113th of 178
Design Elo
Speed
43
192nd fastest
Latency
714ms
first token
Input price
$0.400
146th cheapest
Context
262K

How it compares

Smarter than37%
of all ranked models
Faster than35%
of all ranked models
Cheaper than51%
of all ranked models

Overview

Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window. Devstral 2 supports exploring...

Benchmarks

independent · via OpenRouter
Artificial Analysis36th percentile
Intelligence Index
22.0
Coding Index
23.7
Agentic Index
21.9
GPQA Diamond
59%
Humanity's Last Exam
4%
SciCode
33%
Tau²-Bench (agentic)
25%

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
Mistral$0.400$2.0099.8%

Specifications

Context window262K
Max output
Knowledge cutoff
Input modalitiestext, file
Output modalitiestext
Prompt caching
Cache read price$0.040/M
ModeratedNo

Devstral 2 2512 FAQ

How much does Devstral 2 2512 cost?

Devstral 2 2512 costs $0.400 per million input tokens and $2.00 per million output tokens via OpenRouter, making it 146th cheapest of 298 paid models.

How smart is Devstral 2 2512?

Devstral 2 2512 scores 22.0 on the Artificial Analysis Intelligence Index, ranking 113th of 178 benchmarked models, with a GPQA Diamond score of 59%.

How fast is Devstral 2 2512?

Devstral 2 2512 generates around 43 tokens per second with 714ms time-to-first-token (p50), the 192nd fastest tracked model.

What is Devstral 2 2512's context window?

Devstral 2 2512 supports a 262K-token context window. It accepts text, file input.

Compare head-to-head