Model board / Mistral Small 4

Mistral Small 4

Name: Mistral Small 4
Brand: Mistral
Price: 0.6 USD

latest Mistral

Verified 2026-10-15 against the pages listed below. Prices are USD per 1M tokens, list rate — no committed-use, enterprise or volume discount.

Input

$0.15

Cached input

$0.015

Cache write

—

Output

$0.6

Context window

256,000

Max output

—

Weights

open

Caching

unverified

cached input $0.015 per 1M, printed per model on docs.mistral.ai/inference/pricing — the third Mistral pricing page, added to this row's sources on 2026-09-03. It is the −90% discount the other two pages describe only in prose

window stated on the model's own page; max output not stated

cached input bills at 10% of input in 64-token blocks, so a prompt under 64 tokens never hits; a prompt_cache_key raises the chance of a hit but does not guarantee one. Whether caching applies without the key, and the TTL, are not stated

Licence: Apache-2.0 (v26.03)

Watch out

Hybrid instruct/reasoning/coding model — pin the mode or output length varies

What we would use it for

Our most-reached-for fine-tuning base — cheap, small, open lineage

What this has cost, week by week

Output price at each refresh since 2026-07-30. A vendor change is the provider changing its price. Our correction means this board changed what it prints and the provider's charge did not move — the two are never merged, because publishing the second as the first would be a price rise that never happened.

07-30 $0.6→08-06 $0.6→08-14 $0.6→08-20 $0.6→08-27 $0.6→09-03 $0.6→09-10 $0.6→09-17 $0.6→09-24 $0.6→10-01 $0.6→10-08 $0.6→10-15 $0.6

our correction2026-08-06 · weights — As above — Apache-2.0 (v26.03) confirmed against the models doc.

newly stated2026-09-03 · cached_in — $0.015 from the same newly cited page. No change to what Mistral charges.

newly stated2026-09-24 · window — 256K window from Mistral's per-model page, newly cited.

Where these numbers came from

No number on this page was written that was not read from one of those pages on 2026-10-15. If one is wrong, send us the source URL and it is corrected in the next weekly refresh, with a changelog line saying what was wrong — contact@naderu.com.

Jobs this is one of our picks for

Translation — Language pairs, with terminology you control (translation)
Content writing — Drafting and rewriting to a house style (content-writing)
Extraction — Classify or pull fields out of a corpus (extraction)
Content moderation — Classify against a policy (content-moderation)

Discussed in Naderu Weekly

The same weights, twice — 2026-08-06

The rest of the board

This row sits alongside 69 others on the model board, refreshed every week. The same data as JSON, and the terms for using it, are at /board-data.