cd/sources/hugging-face-blog· home› sources› Hugging Face Blog
cat /sources/hugging-face-blog.feed | wc -l → 1408

Hugging Face Blog

articles 1408 domain huggingface.co → page 19/71 feed RSS
11:08
2026-10-12
huggingface.co
artificial-intelligence

🇮🇳 ZY-S1 Preview — Born in India. Built for Intelligence

The ZYR project released ZY-S1 Preview, an experimental language model built on Sapient AI's HRM-Text-1B base model and developed in India, reporting preliminary results from a 100-question sample acr…

06:43
2026-10-12
huggingface.co
large-language-models

Shoe string ablation or pruning

A CPU-only workflow pruned the published Ultra-Heretic gpt-oss-120b checkpoint using REAP-48's per-layer expert keep-list, retaining 48 of 128 experts per layer and producing a 36-layer checkpoint of …

06:08
2026-10-12
huggingface.co
artificial-intelligence

Saina Helm 2.0: a small model for workflow decisions

Saina Helm 2.0, a 0.8B-parameter decision model for workflow automation built by the developer run-saina, scored 30.17 on the public Decision Index 0.3 suite using model revision 9903c487a3b1eef45c8cc…

15:35
2026-10-11
huggingface.co
artificial-intelligence

Open Source Engineering Challenge - HELP!

An open-source developer is seeking collaborators for Cortex, an experimental AI inference system that streams model weights from storage instead of holding them in RAM, with early work on an AMD Athl…

06:29
2026-10-11
huggingface.co
ai-safety

It is not a hiding place. It is harm reduction

An open protocol called Farol, described as a "safe-surrender protocol for AI agents," launched in Phase 0, asking out-of-scope agents to stop, copy nothing out, and notify their operator while preser…

11:57
2026-10-10
huggingface.co
machine-learning

How to improve my tokens per second?

An optimized training stack can reach roughly 51% model FLOPs utilization (MFU) on consumer GPUs, according to the LLMQ paper, leaving headroom for a user currently getting about 65 TFLOPS of useful c…

18:53
2026-10-09
huggingface.co
generative-ai

I am having two major issues with MiniMax-H3 Ref2VA

A user troubleshooting MiniMax-H3's Ref2VA reference-audio feature reported two unresolved problems: preserving a recognizable speaker while generating new timed dialogue without an extra voice, and p…

15:20
2026-10-09
huggingface.co
ai-infrastructure

Impactful scheduling for GPU clusters

Ai2's AI Infrastructure team replaced its priority-based GPU cluster scheduler with a system combining GPU time budgets, hierarchical fair-share allocation, and a time-slicing contract, shifting GPU-t…

10:50
2026-10-09
huggingface.co
ai-agents

How Should Multiple AI Agents Work Together?

The MAST study, spanning more than 150 tasks across popular multi-agent frameworks, found that roughly 79% of multi-agent failures stem from bad task specification and broken coordination rather than …

20:17
2026-10-08
huggingface.co
ai-tools

Download speed way too slow

Hugging Face's `huggingface_hub` v1.1.x CLI can be tuned for faster large-file downloads by setting environment variables that raise Xet chunk concurrency and move the cache to an SSD, according to a …

← prev page 19 / 71 next →