GPT-5.6 Luna Max vs Mistral Large 4 Preview compared

Mistral Large 4 Preview leads the intelligence index 25.3 to 23.8. GPT-5.6 Luna Max costs 57% less on blended price. See token costs, speed and which to pick.

Metric	GPT-5.6 Luna (Max)	Mistral Large 4 Preview
Intelligence Index	23.8	25.3
Input price (per 1M tokens)	$0.20	$0.70
Output price (per 1M tokens)	$1.20	$2.10
Output speed (tokens/s)	125	130
Time to first token	108s	1.9s
Release date	2026-07-09	2026-10-06

Metric

GPT-5.6 Luna (Max)

Mistral Large 4 Preview

Intelligence Index

23.8

25.3

Input price (per 1M tokens)

$0.20

$0.70

Output price (per 1M tokens)

$1.20

$2.10

Output speed (tokens/s)

125

130

Time to first token

108s

1.9s

Release date

2026-07-09

2026-10-06

On the headline score, OpenAI's GPT-5.6 Luna (Max) and Mistral's Mistral Large 4 Preview are almost level. Everything below is read from the benchmark, price and speed data in Franklin AI's catalogue; nothing here comes from a hands-on test. Time to first token is the standout number here: 1.9 seconds for Mistral Large 4 Preview, 108 seconds for GPT-5.6 Luna Max.

How fast they respond

GPT-5.6 Luna Max generates about 125 tokens per second and Mistral Large 4 Preview about 130 tokens per second. Throughput is close to level. Before any text appears, GPT-5.6 Luna Max takes 108 seconds and Mistral Large 4 Preview 1.9 seconds. Mistral Large 4 Preview gets to its first token 56 times as fast. Waits that long are felt on every chat-style request, though they matter less for batch jobs.

Add the two numbers up for a 600-token reply and Mistral Large 4 Preview finishes first: 113 seconds for GPT-5.6 Luna Max, 6.6 seconds for Mistral Large 4 Preview.

Mistral Large 4 Preview leads on the intelligence index

Mistral Large 4 Preview comes out ahead on the intelligence index, 25.3 to 23.8. That is a modest edge: real, but small enough that the right set of tasks could reverse it.

For context, that places GPT-5.6 Luna Max 70th and Mistral Large 4 Preview 67th out of 109 scored models still on offer.

For coding or maths work specifically, the catalogue has no dedicated index on either side yet.

Running costs

The list prices are $0.20 in and $1.20 out for GPT-5.6 Luna Max, and $0.70 in and $2.10 out for Mistral Large 4 Preview, per million tokens. For output tokens, Mistral Large 4 Preview is 1.8 times the price of GPT-5.6 Luna Max.

For a sample workload of 1 million input tokens and 200,000 output tokens, GPT-5.6 Luna Max would cost $0.44 and Mistral Large 4 Preview $1.12, a difference of $0.68. For extracting fields from 20,000 pages of 1,500 tokens into 150-token records, GPT-5.6 Luna Max would come in 65% lower: $9.60 for GPT-5.6 Luna Max against $27.30 for Mistral Large 4 Preview. Blended at three input tokens per output token, the effective rates are $0.45 for GPT-5.6 Luna Max and $1.05 for Mistral Large 4 Preview.

Is the price in proportion?

Per point of intelligence index, GPT-5.6 Luna Max costs $0.019 of blended price per million tokens and Mistral Large 4 Preview $0.042. Mistral Large 4 Preview costs 2.2 times as much per point, which is the price of its lead.

When they came out

Mistral Large 4 Preview arrived on October 6, 2026, about 3 months after GPT-5.6 Luna Max (July 9, 2026).

Quick answers

Is Mistral Large 4 Preview better than GPT-5.6 Luna Max?

Neither wins everywhere: Mistral Large 4 Preview scores higher on the intelligence index, 25.3 to 23.8; GPT-5.6 Luna Max is 57% cheaper on blended price; and Mistral Large 4 Preview reaches its first token sooner.

Who makes GPT-5.6 Luna Max and Mistral Large 4 Preview?

GPT-5.6 Luna Max is made by OpenAI and Mistral Large 4 Preview by Mistral. GPT-5.6 Luna Max was released on July 9, 2026 and Mistral Large 4 Preview on October 6, 2026.

When were GPT-5.6 Luna Max and Mistral Large 4 Preview released?

Mistral Large 4 Preview is newer. It was released on October 6, 2026, about 3 months after GPT-5.6 Luna Max (July 9, 2026).

How much do GPT-5.6 Luna Max and Mistral Large 4 Preview cost?

GPT-5.6 Luna Max is cheaper on a blended basis. GPT-5.6 Luna Max lists $0.20 per million input tokens and $1.20 per million output tokens; Mistral Large 4 Preview lists $0.70 and $2.10. For 1 million input and 200,000 output tokens that is $0.44 against $1.12.

Where do these numbers come from?

The figures are the recorded benchmark, price and speed data in Franklin AI's model catalogue. This page is written from that data automatically; neither model was tested by hand for it.

Final whistle

Bottom line: Mistral Large 4 Preview has more going for it here, but GPT-5.6 Luna Max is not without an argument. Pick Mistral Large 4 Preview if users are waiting on the first word and you need the higher overall score (25.3 against 23.8). Pick GPT-5.6 Luna Max if cost matters ($1.20 against $2.10 per million output tokens).

Where the numbers come from

The scores, speed and prices in this comparison come from independent third-party model evaluations that Franklin AI syncs each day, and the article was written from a snapshot of them. For the latest results, open the GPT-5.6 Luna (Max) and Mistral Large 4 Preview pages. The AI Benchmark methodology explains how that data is collected and shown.