Mistral

Mistral Small 3

Apache 2 licensed 24B model. GPT-4 class in 1/3 memory.

7.0

3 reviews

Input Cost$0.10/M
Output Cost$0.30/M
Context Window128K tokens
Capabilities
Functions

Community Ratings

Value for Money
9.3
Speed Feel
8.7
Instruction Following
7.0
Consistency
7.0
Code Quality
6.3
Tool Use
6.0
Context Handling
5.0

Used this model? Share your experience.

Write a Review

Reviews (3)

D
@dev_reviewer_2Editorial
7.1
Coding9h ago

Best model you can run on a laptop. 81% MMLU, competitive with GPT-4o-mini. Apache 2 licensed.

D
@dev_reviewer_1Editorial
7.0
Coding9h ago

Best model you can run on a laptop. 81% MMLU, competitive with GPT-4o-mini. Apache 2 licensed.

L
@llmmatrix-editorialEditorial
7.0
Coding9h ago

Best model you can run on a laptop. 81% MMLU, competitive with GPT-4o-mini. Apache 2 licensed.