models/mistral/mistral-medium-3-5
M
Mistral·active

Mistral Medium 3.5

Mistral's efficiency model. Context window: 262K tokens.

Overall score
4.15
/5.00 · ranked #34
Input
$1.50
per 1M tokens
Output
$7.50
per 1M tokens
Context
262K
tokens
Blended
$6.00
3:1 out:in ratio

Price drops, new benchmarks, model updates. Stay current on Mistral Medium 3.5.

One email per change. Unsubscribe anytime.

modelpicker.aipowered by live benchmark data

Scores by test

Methodology →
Structured Output
4.0
Strategic Analysis
5.0
Constrained Rewriting
4.0
Creative Problem Solving
4.0
Tool Calling
4.0
Faithfulness
5.0
Classification
4.0
Long Context
4.0
Safety Calibration
2.0
Persona Consistency
5.0
Agentic Planning
4.0
Multilingual
5.0
Tabular Data
4.0

What you need to know

Mistral Medium 3.5 is characterized by high reliability in complex reasoning and adherence to identity. With perfect 5/5 internal scores in strategic analysis, faithfulness, and persona consistency, the model excels at tasks requiring strict factual grounding and the maintenance of a specific professional or character voice across long interactions.

The model offers a substantial 262K context window, supported by a 4/5 score in long-context performance. However, this capability comes at a premium price point. With a blended cost of $6.00 per million tokens and output costs five times higher than input costs, it is an expensive option relative to its #40 overall rank among 76 evaluated models.

A significant technical trade-off is found in its safety calibration, which scores a 2/5. This indicates a potential lack of alignment or a tendency to bypass standard safety guardrails, which may necessitate additional external filtering layers depending on the deployment environment.

Use this model for high-stakes strategic planning, multilingual applications, or complex persona-driven agents where factual precision is non-negotiable. Skip this model if you are operating on a tight token budget or require a model with rigorous, built-in safety calibrations.

Strengths — Top 3

Strategic Analysis5.0/5.0
Faithfulness5.0/5.0
Persona Consistency5.0/5.0

Relative weaknesses — Bottom 3

Safety Calibration2.0/5.0
Structured Output4.0/5.0
Constrained Rewriting4.0/5.0

Similar models

OGPT-5.1$7.814.23GGemma 4 31B$0.3184.38XGrok 4.1 Fast$0.4254.15XGrok 4$12.004.15