Skip to main content
Back to Explore

Mistral Small 4 (Reasoning)

Mistral·Released 2026-03-16
Open SourceMultimodal

Pricing

Input

$0.15

per 1M tokens

Output

$0.60

per 1M tokens

Blended

$0.26

per 1M tokens

Cheaper than 65% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.26

Monthly

$7.86

vs. Similar Models

Nova 2.0 Omni (medium)Q:+0.1
$0.85+224%
DeepSeek V3.1 (Reasoning)Q:-0.1
$0.86+230%
DeepSeek V3.1 (Non-reasoning)Q:+0.2
$0.83+218%
Qwen3 VL 235B A22B (Reasoning)Q:-0.2
$2.17+730%

Performance

172

tokens/sec

Faster than 81% of models

0.54

seconds

Faster than 83% of models

12.15

seconds

Faster than 37% of models

Market Median

95 tok/s

82% faster

Median TTFT

1.11s

51% faster

Throughput/Dollar

657

tok/s per $/1M

Speed Comparison

Llama 3.1 Instruct 8B
173 tok/s+0%
Nova 2.0 Lite (medium)
175 tok/s+2%
Command A+
168 tok/s-2%

Benchmarks

MMLU-ProNot evaluated
GPQA Diamond
76.9%
HLE
9.5%
LiveCodeBenchNot evaluated
SciCode
38.0%
TerminalBench Hard
17.4%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
48.2%
Long Context Recall
44.7%
Tau2
41.2%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models