Qeravio
Canonical AI event

Llama.cpp adds n_expert_used_max function for model loading.

The n_expert_used_max function was added to handle specific expert layers in models. It is used in the llama_model_base::load_hparams function.

5 Sept 20261 verified claims1 sources1 observations
What happened

The official source reports this update: b10796: src : add n_expert_used_max function. src : add n_expert_used_max function ( #28323 ) src : add n_expert_used_max function With Commit c61b98b ("model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support ( #25444 )") it is now possible for each layer to have a specific number of experts but there are a few checks that need to be updated to handle this upon model loading.

Why it matters

This official update documents a development concerning b10796: src : add n_expert_used_max function. Its practical significance depends on the scope and evidence stated by the source.

What to watch next

Read the official source update and verify its stated scope, evidence and timing before acting on it.

Connected knowledge

Entities affected by this event

Evidence trail

Sources behind the event