Mistral Medium

Mistral·Released 2023-12-11

Open SourceMoEMultimodal

Model Family

Mistral Small 4 (Reasoning)2026-03-16 Mistral Small 4 (Non-reasoning)2026-03-16 Mistral: Mistral Small 42026-03-16 Mistral: Mistral Small Creative2025-12-16 Mistral Large 32025-12-02 Mistral: Mistral Large 3 25122025-12-01 Mistral Medium 3.12025-08-12 Mistral Small 3.22025-06-20

Quality Index

9.0

383rd of 442

Top 87%

Price/1M

$4.09

601st cheapest

1219% above median

Top 89%

Speed

88 tok/s

Top 33%

TTFT

0.37s

Market Position

Mistral MediumMarket Average

Pricing

Input

$2.75

per 1M tokens

Output

$8.10

per 1M tokens

Blended

$4.09

per 1M tokens

Cheaper than 11% of models. Median price is $0.31/1M tokens.

Cost Calculator

Tokens per day1M

100K100M

Daily

$4.09

Monthly

$122.64

vs. Similar Models

GPT-3.5 TurboQ:0.0

$0.75-82%

Mistral Small (Feb '24)Q:0.0

$1.50-63%

Claude 2.0Q:+0.1

$0.00-100%

DeepSeek R1 Distill Qwen 1.5BQ:+0.1

$0.00-100%

Performance

88

tokens/sec

Faster than 67% of models

0.37

seconds

Faster than 55% of models

0.37

seconds

Faster than 56% of models

Market Median

46 tok/s

92% faster

Median TTFT

0.42s

11% faster

Throughput/Dollar

21

tok/s per $/1M

Speed Comparison

Grok 4.20 Beta 0309 (Non-reasoning)

88 tok/s+0%

Llama Nemotron Super 49B v1.5 (Non-reasoning)

88 tok/s+0%

Olmo 3 7B Instruct

87 tok/s-0%

Benchmarks

MMLU-Pro

49.1%

GPQA Diamond

34.9%

HLE

3.4%

LiveCodeBench

9.9%

SciCode

11.8%

TerminalBench HardNot evaluated

MATH-500

40.5%

AIME

3.7%

AIME 2025Not evaluated

IFBenchNot evaluated

Long Context RecallNot evaluated

Tau2Not evaluated

Market AverageTop Score

Open Source

Quick Compare

Similar Models

GPT-3.5 Turbo

OpenAI

Q: 9.0$0.75/1M

Faster: 20%Cheaper: 82%

Mistral Small (Feb '24)

Mistral

Q: 9.0$1.50/1M

Faster: 72%Cheaper: 63%

Llama 3 Instruct 70B