AIbase
EN
Home
AI Models Finder
Magistral Small
M

Magistral Small

Compact model from the Magistral series, focused on efficiency and resource conservation. Features streamlined design capable of providing reliable AI services under limited computational resources. Particularly suitable for resource-constrained environments and cost-sensitive application scenarios, meets basic text processing, simple conversation, and content assistance needs for SMEs.
Intelligence(Medium)
Speed(Medium)
Input Supported Modalities
No
Is Reasoning Model
128,000
Context Window
128,000
Maximum Output Tokens
2025-06-01
Knowledge Cutoff
Go Compare

Pricing

- /M tokens
Input
- /M tokens
Output
￥5.4 /M tokens
Blended Price

Quick Simple Comparison

Input

Output

Devstral
￥0.1
Devstral Medium
￥0.4
Devstral Small (May '25)
￥0.1

Basic Parameters

Magistral SmallTechnical Parameters
Parameter Count
24,000.0M
Context Length
128.00k tokens
Training Data Cutoff
2025-06-01
Open Source Category
Open Weights (Permissive License)
Multimodal Support
Text Only
Throughput
Release Date
2025-06-10
Response Speed
106.37,368 tokens/s

Benchmark Scores

Below is the performance of Magistral Small in various standard benchmark tests. These tests evaluate the model's capabilities in different tasks and domains.
Intelligence Index
54.72
Large Language Model Intelligence Level
Coding Index
37.75
Indicator of AI model performance on coding tasks
Math Index
83.8
Capability indicator in solving mathematical problems, mathematical reasoning, or performing math-related tasks
MMLU Pro
74.6
Massive Multitask Multimodal Understanding - Testing understanding of text, images, audio, and video
GPQA
64.1
Graduate Physics Questions Assessment - Testing advanced physics knowledge with diamond science-level questions
HLE
7.2
The model's comprehensive average score on the Hugging Face Open LLM Leaderboard
LiveCodeBench
51.4
Specific evaluation focused on assessing large language models' ability in real-world code writing and solving programming competition problems
SciCode
24.1
The model's capability in code generation for scientific computing or specific scientific domains
HumanEval
96.3
Score achieved by the AI model on the specific HumanEval benchmark test set
Math 500 Score
96.3
Score on the first 500 larger, more well-known mathematical benchmark tests
AIME Score
71.3
An indicator measuring an AI model's ability to solve high-difficulty mathematical competition problems (specifically AIME level)
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase