Pretty wild what 900M params can do – PaddleOCR VL > SOTA on OmniDocBench v1.0 & v1.5 (text, tables, formulas, charts, reading order)
> Multilingual – 109 languages (Latin, Arabic, Cyrillic, Devanagari, Thai, etc)
> Handles handwriting, historical docs, noisy scans
> Supports
@reach_vb
-
PaddleOCR VL 900M Params Achieves SOTA on OmniDocBench
By
–
-
Multi-Cloud Replication Strategy for AI Infrastructure
By
–
+ multi cloud replication for good measure
-
GPT-5 Outperforms Sonnet for Idea Brainstorming Tasks
By
–
ChatGPT is the way to go! Specially for bouncing ideas – gpt-5 is way more steerable than sonnet
-
1B Parameter Model: 128K Context, Int4 Quantization, Llama 4 Distilled
By
–
chat is this real??? 128K context, int4 quantisation, 1B params, distilled from Llama 4
-
No prompt storage or sharing with providers
By
–
we don’t share/ store any prompts nor do any of our providers
-
Hugging Face Expands Billing Information Transparency
By
–
Yes! we're working toward adding more information here but part of the reason to keep it vague is to ensure we don't get bots messing/ game storage. but, you can see the exact status on http://
hf.co/settings/billi
ng
… too (atleast for Enterprise/ Pro tiers atm) -
Hugging Face addresses storage limits and open research grants
By
–
Hey hey – VB from HF here, what’s up? Yes, we put storage limits in place but these are much much higher than what the median user on the Hub uses. Besides, we’re actively making grants to ensure all open research is supported. The limits are put in place to help fight rampant
-
IDE-based AI chat tool for improved diff visualization workflow
By
–
It’s been a big part of my workflow in the past few weeks – Looking forward to where it’s headed I only wish I could have better IDE based chat w/ it too primarily to visualise the diff better!
-
Using LLMs to Build Quick Disposable Tools Efficiently
By
–
I love LLMs because I can just prompt them to build simple disposable tools that I'd otherwise not – simple mermaid visualiser it even hooked it up to github pages – all with one prompt; ty codex!