Model APIs - 2 entries
Mistral API rate limits
Mistral's official docs, read on October 17, 2026, set rate limits for the Mistral API. For Models configured for Priority Tier (service_tier=auto) on Priority Tier (enterprise account setup), the docs list 99.5% Uptime SLA.
Last verified from docs.mistral.ai. What the docs leave out
- Sources
- 2
- Tables
- 2
- Entries
- 2
- Verified
- Oct 17, 2026
Limits by model, endpoint and plan
Values are copied exactly as the docs show them, units included, in the docs' own order. A dash means the docs list no value for that row.
Models configured for Priority Tier (service_tier=auto)
Priority Tier (enterprise account setup)
| Metric | Priority Tier (enterprise account setup) |
|---|---|
| Rate limits | Custom and model-specific; "can include requests per minute, tokens per minute, or other limits agreed with Mistral" |
| Bursting | Falls back to Standard Tier when Priority Tier limits are exceeded |
| Uptime SLA | 99.5% |
| Pricing multiplier | 1.75x standard list pricing (75% premium on input, output, and cached tokens) |
- Service_tier values: auto (uses Priority when eligible, falls back to Standard) and standard_only (default when omitted). Standard Tier rate limits described as "Usage-based or custom standard limits".
Studio Workflows managed deployments
Free / Pro or Team (PAYG off) / Pay-as-you-go enabled / Enterprise
| Metric | Free / Pro or Team (PAYG off) / Pay-as-you-go enabled / Enterprise |
|---|---|
| Concurrent managed deployments (Free) | 0 |
| Concurrent managed deployments (Pro or Team, pay-as-you-go off) | 0 |
| Concurrent managed deployments (Pay-as-you-go enabled) | 3 |
| Concurrent managed deployments (Enterprise) | 10 |
| Error on exceed | 403 |
- Workflows also enforce "a 2MB limit on workflow inputs, activity inputs, and activity outputs" (payload offloading page).
How to raise your limits
Models configured for Priority Tier (service_tier=auto)
"Contact your account executive or Mistral contact. Share your target models and expected traffic profile. Wait for Mistral to configure your custom Priority Tier rate limits." "If you need more Priority Tier capacity, coordinate with Mistral ahead of time. Capacity increases may require advance notice."
Studio Workflows managed deployments
Enable pay-as-you-go (Admin Panel > Subscription) or move to Enterprise; stop or delete a deployment to free a slot.
What the docs leave out
- Numeric per-model RPS / tokens-per-minute / tokens-per-month values for any tier: Mistral publishes these only in the logged-in console (https://admin.mistral.ai/plateforme/limits); public docs say only that limits vary by subscription tier and model.
- Numeric audio seconds per minute/month and OCR pages per minute: console only.
- A documented automatic tier-advancement ladder (spend-based tiers): none published; the former docs.mistral.ai/deployment/laplateforme/tier/ page returns 404. Raises go through support or plan/PAYG changes.
- Exact included monthly usage amounts per plan beyond the API credit figures on the pricing page.
- Concurrency limit: not published separately; the Workspace doc describes RPS as 'maximum concurrent API requests'
FAQ
What are the Mistral API rate limits?
Mistral's official docs, read on October 17, 2026, set rate limits for the Mistral API. For Models configured for Priority Tier (service_tier=auto) on Priority Tier (enterprise account setup), the docs list 99.5% Uptime SLA.
How do I raise my Mistral API rate limits?
"Contact your account executive or Mistral contact. Share your target models and expected traffic profile. Wait for Mistral to configure your custom Priority Tier rate limits." "If you need more Priority Tier capacity, coordinate with Mistral ahead of time. Capacity increases may require advance notice." Enable pay-as-you-go (Admin Panel > Subscription) or move to Enterprise; stop or delete a deployment to free a slot.
Where do these Mistral API numbers come from?
Every value is copied verbatim from Mistral's official documentation at docs.mistral.ai, read on October 17, 2026. This page lists all 2 entries in 2 tables, each linked to the page it came from.
Related reading
Sources
Last verified . Every value is copied verbatim from the official pages below, read on that date.
Newsletter
Know when limits change
One email per week. New model releases, pricing and limit changes, and what they mean for what you build.