Transformers now runs llama.cpp quants
The Hugging Face transformers library now supports llama.cpp quantization formats directly, making it easier to deploy quantized models without manual conversion.
source: Hugging Face Blog - Sep 22
The Build - Oct 1, 2026: Holo4, DeepSeek-V4.1-Flash, Transformers llama.cpp quants
Thursday build: Holo4 generalist agents, DeepSeek-V4.1-Flash lands on HF, Transformers now runs llama.cpp quants, and more tools for builders.
For Builders
tokenizers v1 lands
New major release of the tokenizers library with encode/decode improvements and scaling benchmarks.
source: Hugging Face Blog - Sep 21
For subscribers only
Startup
lockedDoorDash launches AI agent you can text to order food
HCompany
lockedHolo4: powering generalist computer-use agents
HF Blog
lockedGive Your Coding Agents a Memory You Own
HF Blog
lockedSource-aware verification for MCP agents
Voice AI
lockedModulate raises $25M for voice models and analysis suite
Security
lockedReco raises $55M as AI agent security startups crowd the market
HF Trending
lockedQwen-Image-2.1: 7B text-to-image model dominates HF Spaces
Voice
lockedElevenLabs v4 speech model more expressive and consistent
HF Trending
lockedDeepSeek-V4.1-Flash: 763B image-text-to-text model
Tabular
lockedNVIDIA Kumo Tabular sets new accuracy-efficiency frontier
HF Spaces
lockedExplore Qwen-Image-2.1 on HF Spaces
The full brief, the tools, the GitHub radar and the context stay subscriber-only. Subscribers always read first, in full.
Read the whole thing next time.
Ten stories, the tools, the trends, zero noise - three times a week, before anyone else.
free forever · no spam · unsubscribe in one click.