Magistral Medium 1.2
Analysis Summary
Magistral Medium 1.2 is Mistral's mid-size reasoning model, released September 2025. Its intelligence index of 17.9 is moderate, but its math index of 82.0, livecodebench score of 0.750, and AIME 2025 score of 0.820 place it well above its general intelligence for quantitative and coding tasks. The agentic index of 32.5 and long-context score of 0.513 are the strongest in the Mistral 3 family shown here. Instruction-following at 0.430 is reasonable.
For businesses, this model is a strong fit for mathematical reasoning, structured content generation, quantitative analysis, and mid-complexity agentic tasks. Its long-context performance makes it more useful than the Small variant for document-heavy workflows. It is not a frontier model for general reasoning or complex software engineering, but it punches above its weight on specialist tasks.
At $2.00 input and $5.00 output per million tokens, it is priced in the mid tier. Teams that need reliable mathematical reasoning and structured output at a cost well below frontier models will find it a practical choice, particularly for finance, data analysis, or content workflows with quantitative components.
Assessed July 10, 2026
Editorial notes
Magistral Medium 1.2 from Mistral combines a strong math index, high livecodebench score, and solid instruction-following with a competitive price, making it a capable mid-tier reasoning and content model.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
DFO Verdict
Magistral Medium 1.2 from Mistral combines a strong math index, high livecodebench score, and solid instruction-following with a competitive price, making it a capable mid-tier reasoning and content model.
Benchmark scores
Magenta = intelligence Ā· Ink = technical/agentic Ā· Cyan = content & long-context Ā· Grey = community benchmarks. Data: Artificial Analysis, Hugging Face.
17.9 Intelligence IndexĀ·21.3 Coding IndexĀ·32.5 Agentic IndexĀ·82 Math Index
How Magistral Medium 1.2 compares
Magistral Medium 1.2 ranks #164 of 393 AI models we track for overall intelligence, #96 of 157 for coding, #125 of 300 for agentic tasks. At $2.00 per million input tokens it is cheaper than 17% of comparable models.
Dark bar = input Ā· light bar = output, scaled to the priciest peer.
1M tokens ā 8 full-length novels or ~2,500 pages of business documents in a single request.
Strongest on value. The pulled-in technical corner is the trade-off, and if the shape matters more than the price, this is your model.
Compare shapes side-by-side āPricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $2.00 | $0.002000 |
| Output | $5.00 | $0.005000 |
What would Magistral Medium 1.2 cost your business?
Pick the job that looks most like yours, then fine-tune with the sliders. Estimates update live.
A website chatbot handling around 100 customer conversations a day, a few short messages each.
Full calculator with 612 models ā Price Calculator
These numbers get smaller with the right architecture.
We route routine calls to cheap models and save Magistral Medium 1.2 for the hard ones. Most clients cut their estimate by 60-80%.
Talk to our teamEmbed this ranking
Frequently asked questions about Magistral Medium 1.2
How much does Magistral Medium 1.2 cost?
Magistral Medium 1.2 costs $2.00 per million input tokens and $5.00 per million output tokens.
Is Magistral Medium 1.2 good for coding?
On our coding benchmark index, Magistral Medium 1.2 ranks #96 of 157 models, placing it in the broader range of the field for code generation and debugging.
Who created Magistral Medium 1.2?
Magistral Medium 1.2 is developed by Mistral and was released on September 18, 2025.
