Magistral Medium
Overview
Overview
Trained solely with reinforcement learning on top of Mistral Medium 3, Magistral Medium is a reasoning model that achieves strong performance on complex math and code tasks without relying on distillation from existing reasoning models. The training uses an RLVR framework with modifications to GRPO, enabling improved reasoning ability and multilingual consistency.
Magistral Medium was released on June 10, 2025.
Performance
Timeline
ReleasedUnknown
Knowledge CutoffUnknown
Specifications
Parameters
24.0B
License
Apache 2.0
Training Data
Unknown
Benchmarks
Benchmarks
Magistral Medium Performance Across Datasets
Scores sourced from the model's scorecard, paper, or official blog posts
llm-stats.com - Sat Feb 21 2026Notice missing or incorrect data?Start an Issue discussion→
Pricing
Pricing
Pricing, performance, and capabilities for Magistral Medium across different providers:
No pricing information available for this model.
API Access
API Access Coming Soon
API access for Magistral Medium will be available soon through our gateway.
Recent Posts
Recent Reviews
FAQ
Common questions about Magistral Medium
Magistral Medium was released on June 10, 2025 by Mistral.
Magistral Medium was created by Mistral.
Magistral Medium has 24.0 billion parameters.
Magistral Medium is released under the Apache 2.0 license. This is an open-source/open-weight license.
Magistral Medium has a knowledge cutoff of June 2025. This means the model was trained on data up to this date and may not have information about events after this time.
Yes, Magistral Medium is a multimodal model that can process both text and images as input.