Mistral Large 4 High vs Step 5 Preview compared

Step 5 Preview leads the intelligence index 28.5 to 26.1. Mistral Large 4 High costs 26% less on blended price. Benchmarks, sample bills, speed and a verdict.

Metric	Mistral Large 4 (High)	Step 5 Preview
Intelligence Index	26.1	28.5
Input price (per 1M tokens)	$0.70	$1.00
Output price (per 1M tokens)	$2.10	$2.70
Output speed (tokens/s)	139	151
Time to first token	1.7s	0.91s
Release date	2026-10-29	2026-09-18

Metric

Mistral Large 4 (High)

Step 5 Preview

Intelligence Index

26.1

28.5

Input price (per 1M tokens)

$0.70

$1.00

Output price (per 1M tokens)

$2.10

$2.70

Output speed (tokens/s)

139

151

Time to first token

1.7s

0.91s

Release date

2026-10-29

2026-09-18

If you only looked at the intelligence index, Mistral Large 4 (High) and Step 5 Preview would be hard to tell apart. What follows is built entirely from tracked data, so where the catalogue is silent, this page says so instead of guessing. The headline gap is on the intelligence index, where Step 5 Preview scores 28.5 and Mistral Large 4 High 26.1.

The headline benchmark

On the intelligence index, Mistral Large 4 High scores 26.1 against 28.5 for Step 5 Preview. It is a small margin for an index that averages many evaluations. Put another way, Mistral Large 4 High would need 9% more on this index to draw level.

Across the 109 active models with an intelligence score in the catalogue, Mistral Large 4 High ranks 65th and Step 5 Preview ranks 61st.

Neither model has a separate coding or math index recorded yet, so those narrower comparisons will have to wait for more data.

How fast they respond

Mistral Large 4 High generates about 139 tokens per second and Step 5 Preview about 151 tokens per second. On long outputs, that is a real advantage for Step 5 Preview. The wait for the first token averages 1.7 seconds on Mistral Large 4 High against 0.91 seconds on Step 5 Preview. Step 5 Preview gets to its first token 1.9 times as fast.

Put together, an 800-token answer would take about 7.5 seconds from Mistral Large 4 High and 6.2 seconds from Step 5 Preview, counting the first-token wait.

Cost: Mistral Large 4 High is cheaper

Mistral Large 4 High lists $0.70 per million input tokens and $2.10 per million output tokens; Step 5 Preview lists $1.00 and $2.70. Step 5 Preview costs 1.3 times as much per output token.

Take 1 million input tokens and 200,000 output tokens: that comes to $1.12 on Mistral Large 4 High and $1.54 on Step 5 Preview. A bigger example, a support assistant answering 10,000 conversations of 2,000 input and 500 output tokens each, costs $24.50 on Mistral Large 4 High and $33.50 on Step 5 Preview: a gap of $9.00.

Value for money

Dividing the blended price by the intelligence index gives a rough cost per point: $0.040 for Mistral Large 4 High and $0.050 for Step 5 Preview, per million tokens. So Step 5 Preview buys its higher score at a premium, while Mistral Large 4 High is the value pick, paying 20% less per point.

Generation gap

Step 5 Preview came first, on September 18, 2026; Mistral Large 4 High followed on October 29, 2026. They are effectively contemporaries.

Quick answers

Is Mistral Large 4 High better than Step 5 Preview?

Not across the board: Step 5 Preview scores higher on the intelligence index, 28.5 to 26.1; Mistral Large 4 High is 26% cheaper on blended price; and Step 5 Preview streams output 1.1 times as fast.

Was this comparison tested by hand?

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.

Does Mistral Large 4 High respond faster than Step 5 Preview?

Step 5 Preview generates output faster, at 151 tokens per second against 139 tokens per second. Step 5 Preview reaches its first token sooner (0.91 seconds against 1.7 seconds).

When were Mistral Large 4 High and Step 5 Preview released?

Mistral Large 4 High is newer. It was released on October 29, 2026, 41 days after Step 5 Preview (September 18, 2026).

Is Step 5 Preview cheaper than Mistral Large 4 High?

Mistral Large 4 High is cheaper on a blended basis. Mistral Large 4 High lists $0.70 per million input tokens and $2.10 per million output tokens; Step 5 Preview lists $1.00 and $2.70. For 1 million input and 200,000 output tokens that is $1.12 against $1.54.

Final whistle

Short answer: Step 5 Preview has more going for it here, but Mistral Large 4 High is not without an argument. Pick Mistral Large 4 High if cost matters ($2.10 against $2.70 per million output tokens) and you want the more recent release. Pick Step 5 Preview if you need the higher overall score (28.5 against 26.1) and users are waiting on the first word.

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 Mistral Large 4 (High) and Step 5 Preview pages. The AI Benchmark methodology explains how that data is collected and shown.