Home / Compare / MiniMax M3 vs Mistral Large 3 2512

MiniMax M3 vs Mistral Large 3 2512

For a typical chat reply (2,000 tokens in, 400 out), Mistral Large 3 2512 costs 1.5× as much as MiniMax M3: $1.60 against $1.08 per 1,000 requests at list price on 2026-09-28.

MiniMax M3Mistral Large 3 2512
Input, per 1M tokens$0.30$0.50
Output, per 1M tokens$1.20$1.50
Context window1.05M262K
Classification or routing (600 tokens in, 10 out), per 1,000$0.192$0.315
Chat reply (2,000 in, 400 out), per 1,000$1.08$1.60
Retrieval-augmented answer (8,000 in, 500 out), per 1,000$3.00$4.75
Confidence signalyesno
Tool callingyesyes
BANKING77, correct of 400316not measured

Which one should carry your traffic?

If MiniMax M3 answers your requests as well as Mistral Large 3 2512, moving that traffic saves 32% on chat-shaped requests. Averages cannot tell you whether it does for your requests. MiniMax M3 exposes a confidence signal, so Gimbix can let it answer first and hand unsure cases to Mistral Large 3 2512.

MiniMax M3 profile · Mistral Large 3 2512 profile · Measure it on your traffic