Mistral Small 4
Mistral AI · released 16 Mar 2026 · open weights
- Open weights, self-hostable
Apache-2.0 open weights, 119B total / 6B active MoE with configurable reasoning effort; the cheapest Mistral row in this ledger.
Context window
256K
256,000 tokens
Input price
$0.15
per 1M tokens
Output price
$0.60
per 1M tokens
Blended
$0.26
3:1 in:out mix
Weights
Open
self-hostable
Released
16 Mar 2026
Mistral AI
Long-context / tier pricing — Batch processing halves both halves; cached input saves up to 90%. Listed prices are the base rate for prompts up to ~200K tokens.
Open-weights API prices are representative third-party-hosted rates; self-hosting costs only your own hardware.
Where it ranks
Of 40 models tracked.
Cost to run
#6
of 40
Input price
#3(tied)
of 40
Context window
#36(tied)
of 40
Compare with
The nearest models by blended cost to run.
Put it to work
Specs from BitByteCore's hand-verified Frontier Ledger (v2026-09-23), this row verified 22 Sep 2026 against Mistral AI's official pricing ↗. Model pricing moves fast, so confirm with Mistral AI before you rely on a figure.