Comparison

GPT-4o vs Mistral Large 3

Per input token, Mistral Large 3 runs 80% cheaper than GPT-4o. Mistral Large 3 is 85% cheaper on output tokens. GPT-4o offers a cheaper cached-input tier for repeated context.

FieldGPT-4o
Legacy
Mistral Large 3
ProviderOpenAIMistral AI
Context window128K
Max output
Input / 1M$2.50$0.5080% cheaper
Output / 1M$10.00$1.5085% cheaper
Cached input / 1M$1.25
Batch discount50%
Release date

Worked cost examples

Estimated monthly cost on three common usage shapes

WorkloadGPT-4oMistral Large 3

Chatbot

800 input / 200 output tokens per request, 100,000 requests/month

$400.00$70.00cheaper

RAG pipeline

6,000 input / 700 output tokens per request, 30,000 requests/month

$660.00$121.50cheaper

Batch summarization

20,000 input / 1,500 output tokens per request, 5,000 requests/month

$325.00$61.25cheaper

GPT-4o vs Mistral Large 3 FAQ

Is GPT-4o cheaper than Mistral Large 3?
GPT-4o costs $2.50 per 1M input tokens versus $0.50 for Mistral Large 3 — Mistral Large 3 is 400% cheaper per input token.
Which is cheaper for output tokens, GPT-4o or Mistral Large 3?
GPT-4o charges $10.00 per 1M output tokens; Mistral Large 3 charges $1.50 per 1M output tokens.
Do GPT-4o and Mistral Large 3 offer cache pricing?
GPT-4o discounts cached input to $1.25 per 1M tokens. Mistral Large 3 has no documented cache discount.

Looking for another matchup? See all comparisons.