The trick: Announced Not Shipped

Mistral calls Large 4 open-weight.

The weights are not out until October 27, and an index ranks it below a cheaper rival.

Issue 377 October 202614 receipts3 min

Mistral says Large 4 pushes the frontier of open-weight performance and is state-of-the-art among open models on finance and law.

Before you read on. Your call?

Its own announcement says the weights drop at the end of the month, and thenextweb.com dates that October 27.

The twist

On the Vals Index, Large 4 scores 48.05% at $13.78 per test, behind GLM 5.3 at 53.51% and $7.25.

48.05%Mistral Large 4 score on the Vals Index

53.51%GLM 5.3 score on the same index

$13.78Mistral Large 4 cost per test on the Vals Index

$7.25GLM 5.3 cost per test on the Vals Index

There’s more to this story.

Membership opens the full investigation, the strongest counterargument and what to do with what you’ve learned.

Start your 30-day free trial →

First membership: 30 days free, then A$89 a year. One introductory trial per customer. Card required; renews annually until cancelled. Cancel before the trial ends to avoid the first charge. Already a member? Sign in

The trick has a name

We call it Announced Not Shipped: claimed, never released. You'll see it again. Learn to spot it →

Say this in tomorrow's meeting“Mistral Large 4 is called open-weight, but the weights are not out until October 27, and on the Vals Index it scores below GLM 5.3 at about twice the cost per test.”

Receipts

Supports mistral.ai: ML4 pushes the frontier of open-weight performance.
Context mistral.ai: Weights drop end of this month.
Supports mistral.ai: It already achieves performance competitive with the strongest open-source models globally
Supports mistral.ai: we find it to be state-of-the-art among open models.
Supports mistral.ai: it reasons more precisely and with more structure than GLM-5.3
Context mistral.ai: 1 trillion-parameter natively multimodal model with 49 billion active parameters
Supports docs.mistral.ai: Mistral Large 4 is a state-of-the-art, open-weight, general-purpose multimodal model
Context docs.mistral.ai: It features 52B active parameters and 1.05T total parameters
Context thenextweb.com: will be available on October 27.
Context thenextweb.com: The benchmark results are preliminary, and the company expects them to change before the weights are released.
Refutes thenextweb.com: though it has not shared figures to back that up
Refutes vals.ai: 32 Mistral Large 4 48.05% $13.78 $1.36 / $4.18
Refutes vals.ai: 16 GLM 5.3 53.51% $7.25 $1.4 / $4.4
Context vals.ai: agentic model performance across finance, coding, legal, and tax tasks

Open the Receipts Pack → What each source proves, every figure traced, and what would change our verdict.