Mistral Large 3 2512

Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.

Share

Model details

Context window262,144 tokens
Max completion size68 tokens
Prompt cost / 1K tokens$0.0000005
Completion cost / 1K tokens$0.0000015
Accepts
Produces

Benchmark performance

Overall

72
score
13th
placement

Cost

95
score
6th
placement

Logic

84
score
5th
placement

Speed

87
score
13th
placement

Scoring

26
score
18th
placement

Tool Use

45
score
5th
placement

Hallucination

80
score
8th
placement

Classification

29
score
3rd
placement

Structured Output

67
score
5th
placement

Plans and pricing

Usage pricing
Prompt$0.0000005
Completion$0.0000015
RequestFREE
ImageFREE
Web SearchFREE
Internal ReasoningFREE

Best Overall scoring LLMs

88
score
1st
placement
86
score
2nd
placement
84
score
3rd
placement
82
score
4th
placement
82
score
4th
placement
80
score
5th
placement