Magistral Medium 1.2
Available
2025-09-18
32,000 tokens
Mistral 开发的 AI 模型，具有出色的推理能力，适合多种 AI 应用场景。
Supported modalities:TextCode
Quick Overview
Text Generation3/10
Code Generation2/10
Reasoning8/10
Multimodal2/10
Benchmark Results
Scores from leading benchmark suites.
27.0%
artificial-analysis-intelligence
21.7%
artificial-analysis-coding
82.0%
artificial-analysis-math
Performance Metrics
Latency and throughput performance.
1000ms
P50 Latency
47.886
tokens/sec
Dive Deeper
Explore detailed analyses by topic.
Pricing Info
Feature support
Streaming
Tool calling
JSON mode