Skip to main content
Back to Explore

Mistral Large 3

Mistral·Released 2025-12-02
Open SourceMoEMultimodal

Pricing

Input

$0.50

per 1M tokens

Output

$1.50

per 1M tokens

Blended

$0.75

per 1M tokens

Cheaper than 44% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.75

Monthly

$22.50

vs. Similar Models

Qwen3.5 4B (Non-reasoning)Q:+0.1
$0.06-92%
Qwen3 30B A3B 2507 (Reasoning)Q:-0.1
$0.67-10%
DeepSeek V3 0324Q:-0.2
$1.21+61%
INTELLECT-3Q:-0.3
$0.43-43%

Performance

56

tokens/sec

Faster than 25% of models

0.70

seconds

Faster than 71% of models

0.70

seconds

Faster than 79% of models

Market Median

95 tok/s

41% slower

Median TTFT

1.11s

37% faster

Throughput/Dollar

74

tok/s per $/1M

Speed Comparison

Pixtral Large
56 tok/s-0%
Qwen2.5 Instruct 72B
56 tok/s-0%
Qwen3.5 Omni Plus
55 tok/s-1%

Benchmarks

MMLU-Pro
80.7%
GPQA Diamond
68.0%
HLE
4.1%
LiveCodeBench
46.5%
SciCode
36.2%
TerminalBench Hard
15.9%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025
38.0%
IFBench
36.2%
Long Context Recall
34.7%
Tau2
24.6%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models