Mistral Medium 3.1 API Pricing
Mistral Medium 3.1 was retired by Mistral on August 31, 2026 - the mistral-medium-2508 model id no longer answers. Mistral names Mistral Medium 3.5 as the replacement. Its last listed rates were $1.50/M input and $7.50/M output, kept here for invoice checks and migration math. Checked against Mistral's deprecated and retired models table.
Run the numbers.
Live calculator pre-loaded with Mistral Medium 3.1 rates. Tweak spend, output mix, or cache assumptions to compare it with sibling models.
Real-world presets.
Invoice extraction
Support assistant
Knowledge base answer
Image QA review
Counts use a chars-per-token calibration measured on the vendor's own published tokenizer (mistralai/Mistral-Nemo-Instruct-2407, 2026-06-10). English prose is typically within a few percent; code and non-Latin scripts tokenize heavier. For billing-exact counts use the vendor's count-tokens API.
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| Mistral Medium 3.1 Current | $1.50 | $7.50 | $1.98 agentic 92/8 | documented elsewhere | Mistral production workloads |
| Mistral Large 3 | $0.50 cache $0.05 | $1.50 | $0.241 −8.2× | documented elsewhere — not on pricing page | Mistral production workloads |
| Mistral Small 4 | $0.15 cache $0.015 | $0.60 | $0.0842 −24× | 256K | Low-cost multimodal chat |
| Magistral Medium | $2.00 | $5.00 | $2.24 retired | documented elsewhere — not on pricing page | Transparent reasoning |
| Magistral Small | $0.50 | $1.50 | $0.58 retired | documented elsewhere | Transparent reasoning |
| Ministral 3 14B | $0.20 cache $0.02 | $0.20 | $0.0642 −31× | 128K | Mistral production workloads |
| Ministral 3 3B | $0.10 cache $0.01 | $0.10 | $0.0321 −62× | 128K | Mistral production workloads |
| DeepSeek V4.1 Flash | $0.30 cache $0.006 | $1.20 | $0.15 −13× | 1M | Budget coding and bulk reasoning |
Frequently asked.
Practical pricing questions, separated from calculator assumptions.
Q · 01 Can I still call Mistral Medium 3.1? +
mistral-medium-2508 on August 31, 2026, and requests to it now fail. Mistral names Mistral Medium 3.5 as the replacement. The rates on this page are the last ones Mistral listed, kept for reading old invoices and for estimating a migration.Q · 02 What is Mistral Medium 3.1 priced at? +
$1.5/M input and $7.5/M output. The page stores USD per-million-token baseline pricing from mistral.ai.Q · 03 Does this page include higher context pricing tiers? +
Q · 04 Is prompt caching priced separately? +
$1.5/M.Q · 05 How is the effective price calculated? +
$1.98/M with only documented cache discounts included.