AI Model ComparisonAI Model Comparison

Devstral Small 2

Available
Mistral logo
Mistral
2025-12-09
32,000 tokens

Mistral 开发的 AI 模型，适合多种 AI 应用场景。

Try Now
Supported modalities:TextCode
Quick Overview
Text Generation2/10
Code Generation2/10
Reasoning3/10
Multimodal2/10
Benchmark Results
Scores from leading benchmark suites.
19.3%
artificial-analysis-intelligence
20.7%
artificial-analysis-coding
34.3%
artificial-analysis-math
Performance Metrics
Latency and throughput performance.
1000ms
P50 Latency
197.055
tokens/sec
Dive Deeper
Explore detailed analyses by topic.
Pricing Details
Billing structure and cost analysis
Capability Breakdown
Feature highlights and use cases
Benchmarks
Detailed evaluation data
Pricing Info
Input (1K tokens)$0.01
Output (1K tokens)$0.03
View detailed pricing
Feature support
Streaming
Tool calling
JSON mode
Devstral Small 2 Model Review & Specs