Mistral Rate Limits by Tier
What are Mistral's API rate limits?
Mistral's entry tier (Experiment) allows an unpublished number of requests/min and an unpublished number of tokens/min for documented default/model family. Limits scale up through Production as cumulative spend and account age increase — see the full table below, verified 2026-08-14.
Limits by tier
| Tier | Qualification | Model class | RPM | TPM | RPD | Concurrent |
|---|---|---|---|---|---|---|
| Experiment | Free experimentation account | documented default/model family | — | — | — | — |
| Production | Paid account; model/account limits apply | documented default/model family | — | — | — | — |
— means not documented by Mistral, never a guess.
What this means for your workload
Classification at volume: 116 calls/min and 60,320 tokens/min at the production profile.
This provider publishes no numeric cap for this workload; check the account console before launch.
Response headers
retry-after | Seconds to wait before retrying, when supplied with a 429 |
rate-limit response headers | Provider-specific remaining and reset counters when documented |
When you exceed the limit
Mistral returns HTTP 429.
Reduce concurrency and retry with backoff.
FAQ
What happens when I exceed Mistral's rate limit?
Mistral returns HTTP 429 with a retry-after header telling you how long to wait. Reduce concurrency and retry with backoff.
How do I request a rate limit increase on Mistral?
Request one from the account dashboard: https://console.mistral.ai
Batch 48 · mistral rate-limit decision and evidence modules. Surface verification: 2026-08-14. These are dated, route-local fixtures, not live quota claims.
Mistral effective-limit discovery ledger
Frozen mistral evidence board: organization default · workspace cap · second workspace · service account · free/developer · enterprise override. Formula / decision rule: effective minimum = min(known account/workspace/endpoint values); unpublished values remain Unavailable
| Frozen fixture | Inputs and observation | Formula / boundary | State |
|---|---|---|---|
batch48-mistral-m1-r1organization default · workspace cap | principal/account=joined; endpoint=joined; model/pool=joined; bucket/window=joined; observed=2026-08-14 The primary identity and product joins are present for this frozen fixture. | effective minimum = min(known account/workspace/endpoint values); unpublished values remain Unavailable A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | PASS WITH SCOPE — frozen observation only. |
batch48-mistral-m1-r2second workspace · service account | request/response/error=joined; limit/remaining/reset=observed; retry identity=joined; region=joined The response evidence is usable only for the named bucket and time window; no neighboring provider is borrowed. | effective minimum = min(known account/workspace/endpoint values); unpublished values remain Unavailable A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | PASS WITH LIMIT — dated evidence only. |
batch48-mistral-m1-r3free/developer · enterprise override | workload=joined; prompt/output reserve=explicit; shared pool=declared; unknown=Unavailable; owner=platform Admission or recovery is withheld where the account-specific value or side-effect identity is absent. | effective minimum = min(known account/workspace/endpoint values); unpublished values remain Unavailable A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | UNAVAILABLE — live owner evidence required. |
Provenance: Frozen Batch 48 mistral fixture: module 1 observation and calculation. First-party evidence checked 2026-08-14; unresolved joins fail closed. Surface verification date: 2026-08-14. Mistral API docs (first-party source)
Mistral workload admission worksheet
Frozen mistral evidence board: short chat · 32K extraction · long generation · embeddings batch · OCR/document · audio · 50-worker burst. Formula / decision rule: numerical admission requires a live account limit; known buckets only
| Frozen fixture | Inputs and observation | Formula / boundary | State |
|---|---|---|---|
batch48-mistral-m2-r1short chat · 32K extraction | principal/account=joined; endpoint=joined; model/pool=joined; bucket/window=joined; observed=2026-08-14 The primary identity and product joins are present for this frozen fixture. | numerical admission requires a live account limit; known buckets only A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | PASS WITH SCOPE — frozen observation only. |
batch48-mistral-m2-r2long generation · embeddings batch | request/response/error=joined; limit/remaining/reset=observed; retry identity=joined; region=joined The response evidence is usable only for the named bucket and time window; no neighboring provider is borrowed. | numerical admission requires a live account limit; known buckets only A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | PASS WITH LIMIT — dated evidence only. |
batch48-mistral-m2-r3OCR/document · audio · 50-worker burst | workload=joined; prompt/output reserve=explicit; shared pool=declared; unknown=Unavailable; owner=platform Admission or recovery is withheld where the account-specific value or side-effect identity is absent. | numerical admission requires a live account limit; known buckets only A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | UNAVAILABLE — live owner evidence required. |
Provenance: Frozen Batch 48 mistral fixture: module 2 observation and calculation. First-party evidence checked 2026-08-14; unresolved joins fail closed. Surface verification date: 2026-08-14. Mistral API docs (first-party source)
Mistral 429 and increase-path runbook
Frozen mistral evidence board: rate throttle · workspace spend cap · organization spend cap · endpoint overload · invalid plan · partial stream · repeated retry. Formula / decision rule: retry = provider-authorized transient ∧ bounded backoff ∧ duplicate-safe effect
| Frozen fixture | Inputs and observation | Formula / boundary | State |
|---|---|---|---|
batch48-mistral-m3-r1rate throttle · workspace spend cap | principal/account=joined; endpoint=joined; model/pool=joined; bucket/window=joined; observed=2026-08-14 The primary identity and product joins are present for this frozen fixture. | retry = provider-authorized transient ∧ bounded backoff ∧ duplicate-safe effect A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | PASS WITH SCOPE — frozen observation only. |
batch48-mistral-m3-r2organization spend cap · endpoint overload | request/response/error=joined; limit/remaining/reset=observed; retry identity=joined; region=joined The response evidence is usable only for the named bucket and time window; no neighboring provider is borrowed. | retry = provider-authorized transient ∧ bounded backoff ∧ duplicate-safe effect A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | PASS WITH LIMIT — dated evidence only. |
batch48-mistral-m3-r3invalid plan · partial stream · repeated retry | workload=joined; prompt/output reserve=explicit; shared pool=declared; unknown=Unavailable; owner=platform Admission or recovery is withheld where the account-specific value or side-effect identity is absent. | retry = provider-authorized transient ∧ bounded backoff ∧ duplicate-safe effect A missing provider, product, account, endpoint, model, bucket, or time-window join fails closed; no value is copied across routes. | UNAVAILABLE — live owner evidence required. |
Provenance: Frozen Batch 48 mistral fixture: module 3 observation and calculation. First-party evidence checked 2026-08-14; unresolved joins fail closed. Surface verification date: 2026-08-14. Mistral API docs (first-party source)