Qeravio
Canonical AI event

Llama.cpp updates kv-cells to improve token lookup efficiency.

The update improves throughput by 4.9% on Qwen3.8-Flash-Next UD-Q4_K_XL.

2 Sept 20261 verified claims1 sources1 observations
What happened

The official source reports this update: b10750: kv-cells: look up the n-gram history in the sequence position index. kv-cells: look up the n-gram history in the sequence position index ( #28040 ) get_prev_tokens() rebuilt a (seq, pos) -> token hash map on every ubatch by walking all used cells, while llama_kv_cells already keeps an ordered index of the positions of each sequence in seq_pos, updated on every cell mutation to serve seq_pos_min() and seq_pos_max().

Why it matters

This official update documents a development concerning b10750: kv-cells: look up the n-gram history in the sequence position index. 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