PulseAugur
EN
LIVE 00:39:54

llama.cpp adds Gemma 4 audio stream support in b12237 release

The llama.cpp project has released version b12237, which includes support for Gemma 4's input audio stream. This update addresses an issue where real-time embeddings were incorrectly forced on for audio generation, causing chat models with audio streams to abort. The fix ensures that embeddings are only enabled when necessary for audio generation. AI

IMPACT Improves audio streaming capabilities for models like Gemma 4 within the llama.cpp framework.

RANK_REASON This is a software update for a specific tool, not a frontier model release.

Read on llama.cpp — Releases →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

llama.cpp adds Gemma 4 audio stream support in b12237 release

How we ranked this

Signal score
19 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
This is a software update for a specific tool, not a frontier model release.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. llama.cpp — Releases TIER_1 English(EN) · ngxson ·

    b12237: mtmd: support gemma 4 input audio stream (#31965)

    <ul> <li> <p>mtmd: support gemma 4 input audio stream</p> </li> <li> <p>use 480ms step</p> </li> <li> <p>server: enable realtime embeddings only for audio generation</p> </li> </ul> <p>The realtime slot forced embeddings on, which marks every row of a<br /> batch as an output. On…