Back to Model Tracker
Mistral Active

Mistral Medium 3.5

Mistral’s current frontier-class multimodal model, positioned for agentic and coding use with a 256K context window.

Mistral Medium Updated May 20, 2026
Post Reboot Practical Scores

Capability profile

84.0
Reasoning 84.0
Coding 88.0
Writing 81.0
Speed 86.0
Cost efficiency 80.0
Image understanding 82.0
Pricing
Input
$1.5 / 1M tokens
Output
$7.5 / 1M tokens
Cached input
See Mistral pricing
Context
256K tokens
Free tier
No / unknown
API
Available
Subscription
Mistral API / La Plateforme
Open-source status
Closed
Best Use Cases
Best for
  • Developers who want a capable European API option for coding and agent-style product work.
Best workflows
  • General copilots
  • coding tasks
  • multimodal assistance
  • and product-integrated AI features.
Weakest at
  • Maximum public mindshare and benchmark visibility compared with the biggest US labs.
Not ideal for
  • Projects that only want the most benchmark-famous flagship available.
Post Reboot Take

Mistral’s role in the market is strategic as much as technical. It gives serious teams another commercial frontier path that is not just a copy of the biggest platforms.

Mistral’s current frontier-class multimodal model, positioned for agentic and coding use with a 256K context window.

Post Reboot take

Mistral’s role in the market is strategic as much as technical. It gives serious teams another commercial frontier path that is not just a copy of the biggest platforms.

Strengths

Best for: Developers who want a capable European API option for coding and agent-style product work.

Weakest at: Maximum public mindshare and benchmark visibility compared with the biggest US labs.

Best workflows: General copilots, coding tasks, multimodal assistance, and product-integrated AI features.

Example use cases: Internal assistants, coding support, document/image analysis, and European-hosted workflows.

Sources used for this profile