Magistral Medium 1.2 vs Magistral Small 1.2

Comparing 2 AI models · 5 benchmarks · Mistral

Most Affordable
Mistral logo
Magistral Small 1.2
$0.50/1M
Highest Intelligence
Mistral logo
Magistral Medium 1.2
73.9% GPQA
Best for Coding
Mistral logo
Magistral Medium 1.2
42.3 Coding Index
Price Difference
4.0x
input cost range

Composite Indices

Intelligence, Coding, Math

Standard Benchmarks

Academic and industry benchmarks

Benchmark Winners

5 tests
Mistral logo

Magistral Medium 1.2

5
  • GPQA
  • MMLU Pro
  • HLE
  • LiveCodeBench
  • AIME 2025
Mistral logo

Magistral Small 1.2

0

No clear wins

Metric
Mistral logo Magistral Medium 1.2
Mistral
Mistral logo Magistral Small 1.2
Mistral
Pricing
Per 1M tokens
Input Cost $2.00/1M $0.50/1M
Output Cost $5.00/1M $1.50/1M
Blended Cost 3:1 input/output ratio
$2.75/1M $0.75/1M
Specifications
Organization Model creator
Mistral Mistral
Release Date Launch date
Sep 18, 2025 Sep 17, 2025
Performance & Speed
Throughput Output speed
95.0 tok/s 194.0 tok/s
Time to First Token (TTFT) Initial response delay
456ms 352ms
Latency Time to first answer token
21506ms 10664ms
Composite Indices
Intelligence Index Overall reasoning capability
52.0 43.0
Coding Index Programming ability
42.3 37.2
Math Index Mathematical reasoning
82.0 80.3
Standard Benchmarks
GPQA Graduate-level reasoning
73.9% 66.3%
MMLU Pro Advanced knowledge
81.5%