PulseAugur
EN
LIVE 20:34:05

llama.cpp b11813 release updates Pixtral 12B tokenizer, integrates with new models

The llama.cpp project has released version b11813, which includes updates to its pre-tokenizer for the Pixtral 12B model, moving away from the Llama 3 tokenizer. This release also details compatibility and build options across various operating systems including macOS, Linux, Android, and Windows, with specific configurations for openEuler. Additionally, the Hugging Face model prism-ml/Ternary-Bonsai-2-9B-gguf is highlighted, with extensive instructions provided for its integration with llama.cpp and other popular inference tools like LM Studio, Jan, vLLM, and Ollama. AI

IMPACT Facilitates broader adoption of new models by improving compatibility and integration with popular inference tools.

RANK_REASON This cluster details updates to a software tool (llama.cpp) and the integration of a specific model (prism-ml/Ternary-Bonsai-2-9B-gguf) with various inference platforms.

Read on llama.cpp — Releases →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

llama.cpp b11813 release updates Pixtral 12B tokenizer, integrates with new models

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
This cluster details updates to a software tool (llama.cpp) and the integration of a specific model (prism-ml/Ternary-Bonsai-2-9B-gguf) with various inference platforms.
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
model release, product
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
57 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [2]

  1. llama.cpp — Releases TIER_1 (SO) · github-actions[bot] ·

    b11813

    <details open=""> <p>vocab : use Tekken pre-tokenizer for pixtral rather than the Llama 3 one (<a class="issue-link js-issue-link" href="https://github.com/ggml-org/llama.cpp/pull/30960">#30960</a>)</p> </details> <p><strong>Website:</strong></p> <ul> <li><a href="https://llama.a…

  2. Hugging Face Trending Models TIER_1 Bahasa(ID) · prism-ml ·

    prism-ml/Ternary-Bonsai-2-9B-gguf

    image-text-to-text · 87,480 downloads · 109 likes