Skip to main content

Mistral

Official API

Mistral makes these models and serves them through its own API, at the reference price every other provider is compared with.

Visit mistral.ai
Models21
Cheapest for
1 models
Serves
4 model types
Monthly visits—
Models by type

Mistral's API

ModelInputOutputLowest third-partyContext
GLM-5.3$1.40/1M$4.40/1M
from $1.10 at NanoGPT
—1.05M
Mistral Large 4$1.36/1M$4.18/1M
from $2.09 at Opper
$2.09/1MOpper-50%1.05M
GLM-5.3-Flash$0.15/1M$0.50/1M
from $0.20 at APIMart
—1.05M
Ministral 3 3B 2512Only here$0.10/1M$0.10/1M—262K
Mistral OCR 4.1Only here—From$4/1K pages——
Voxtral TTS 2603Only here—$0.016/1K chars——
Ministral 3 8B 2512Only here$0.15/1M$0.15/1M—262K
Ministral 3 14B 2512Only here$0.20/1M$0.20/1M—262K
Mistral Small 4$0.15/1M$0.60/1M$0.60/1MEmpirioLabs AI262K
Voxtral SmallOnly here$0.10/1M$0.40/1M—33K
Codestral 25.08$0.30/1M$0.90/1M$0.90/1MDevPass (LLM Gateway)131K
Mistral Large 3$0.50/1M$1.50/1M$1.50/1MAIHubMix262K
Mistral Medium 3.5 128BCheapest$1.50/1M$7.50/1M—262K
Mistral Moderation 2603Only hereFree/1MFree/1M—131K
Mistral OCR 2512Only here—From$2/1K pages——
Voxtral Mini Transcribe 2602Only here—$0.003/min——
Voxtral Mini Transcribe 3Only here—$0.004/min—33K
Voxtral Mini Transcribe Realtime 2602Only here—$0.006/min——
Voxtral Mini Transcribe Realtime 3Only here—$0.008/min——
Voxtral Small Transcribe 3Only here—$0.004/min—33K
Voxtral TTS 3Only here—$0.013/1K chars——

Mistral at a glance

21 models in total: 11 text, 7 audio, 2 vision, 1 moderation. The cheapest is Ministral 3 3B 2512 at $0.10 /1M. GLM-5.3 has the largest context window, 1.05M tokens. The most searched is GLM-5.3, with about 276K Google searches a month. The newest is Voxtral Mini Transcribe 3, released Oct 20, 2026.