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 2/71 feed RSS
09:25
2026-11-20
huggingface.co
large-language-models

Running local AI models

A practitioner running Ollama on an 8GB M1 MacBook Pro reported that quantized 1B–2B parameter models such as Qwen 0.6B and 1.7B remain usable for narrow tasks like classification, information extract…

05:08
2026-11-20
huggingface.co
large-language-models

GGUF Merge for QWEN, what is the best way that you know of?

A community troubleshooting write-up recommends establishing a working baseline with an existing Qwen GGUF before attempting any weight merges, citing the official Qwen2.5-Coder-1.5B-Instruct GGUF fil…

11:21
2026-11-19
huggingface.co
ai-infrastructure

What's your setup for running models locally?

A Hugging Face forum discussion on local model inference drew advice that a 7B or 8B model at 4-bit quantization runs comfortably on 16GB of unified memory on a MacBook Pro, while anything larger star…

10:51
2026-11-19
huggingface.co
machine-learning

Time to talk about Katana

An independent tester ran a 1,457,856-parameter byte-level decoder on TinyShakespeare for 144 optimizer steps on a Colab NVIDIA L4 with PyTorch 2.14.1+cu130 and found that Avalanche's public mode="non…

04:00
2026-11-19
huggingface.co
ai-agents

Why AI agents can change behavior without a code change

AI agents can change behavior without any code change when a model provider ships an update, because the underlying model is a dependency whose new weights can shift how instructions are interpreted a…

23:15
2026-11-18
huggingface.co
ai-safety

How internal AI assistants leak data through permissions

A Hugging Face forum post warns that internal AI assistants leak sensitive data when the retrieval layer merges documents of different sensitivity levels into a single context window before the LLM pr…

18:53
2026-11-18
huggingface.co
machine-learning

Why your fine-tuning job takes longer than the FLOPs say

A CatBoost regressor trained on 272 production supervised fine-tuning jobs on NVIDIA H200 GPUs predicted end-to-end job duration with R² = 0.98 and a mean absolute error of 1.71 hours (median 0.51 hou…

14:38
2026-11-18
huggingface.co
ai-agents

Architectural Gravity – Cross-Model Agent Control Language

A synthetic six-rule routing policy test using one 4B GGUF model found that the model recovered all named rule relationships in 10 of 12 strict trials (12 of 12 after a narrow `Rule:` label normalizat…

← prev page 2 / 71 next →