benchr
Pricing Guide · Mistral

Mistral Medium 3.5 API Pricing

Detailed token costs, context window limits, benchmark performance, and caching structures for Mistral Medium 3.5.

Last reviewed: June 6, 2026 Sourced from official docs

Mistral Medium 3.5 Cost Breakdown

Dimension Value / Cost
Input Tokens Cost / 1M$1.500
Output Tokens Cost / 1M$7.500
Cached Input Tokens Cost / 1M—
Max Context Window256,000 tokens
SWE-bench Verified Score77.6%
GPQA Diamond Score—

Best suited for:

  • European data residency
    • \n
  • Multimodal + reasoning in one model
    • \n
  • Self-hosted on 4 GPUs

← Back to all pricing guides