Data collection is inevitable in the computer age. But what do #AI chatbots collect, and how do they compare to each other? https://
visualcapitalist.com/ranked-which-a
i-chatbots-collect-the-most-data-about-you/
… via @visualcap #ArtificialIntelligence @jblefevre60 @FrRonconi @CurieuxExplorer @bamitav @enilev @Khulood_Almani @Shi4Tech
LLMS
-
AI Chatbots Data Collection Comparison and Rankings
By
–
-
Quantization Tutorial: Reduce AI Model Operational Costs
By
–
Bonus: We give a full tutorial on Quantization, including how it works, the different techniques, and more. Bonus 2: We have a detailed Google Colab for you in there! Want to dive deeper? Read our blueprint to reduce AI model operational costs with quantization:
-
4-bit quantization reduces inference overhead for resource-constrained deployment
By
–
While the 4‑bit model experiences a 30–45% increase in inference time due to dequantization overhead, the output remains coherent and accurate. This approach makes deploying large models on resource-constrained hardware much more feasible.
-
Llama 3 8B quantized to 4-bit reduces memory usage by 64%
By
–
We converted a full-precision (FP16) Llama 3 8B model into a 4‑bit version using Hugging Face’s bitsandbytes library with the “nf4” configuration. This reduced the model’s memory usage from ~15 GB to ~5.4 GB—a savings of roughly 64%.
-
Character AI tests file uploads with vision
By
–
Character AI is testing file uploads. Seems like it is now powered by a model with vision capabilities. I think they should gust get native image gen tbh
-
DeepSeek v3 0324 Deep Dive Analysis Guide
By
–
Check the deep dive in the new DeepSeek v3 version 0324 of you’re interested in understanding more about this model:
-
DeepSeek V3 and Vibe-Coding LLMs Transform App Development
By
–
The new DeepSite space on Hugging Face is totally insane for vibe-codershttps://t.co/dKsMEDWOLN
— Thomas Wolf (@Thom_Wolf) 30 mars 2025
With the new wave of vibe-coding-optimized LLMs like the latest open-source DeepSeek model (version V3-0324), you can basically prompt out-of-the-box and create any app and game in… pic.twitter.com/JqBFQ74NsVThe new DeepSite space on Hugging Face is totally insane for vibe-coders https://
huggingface.co/spaces/enzostv
s/deepsite
… With the new wave of vibe-coding-optimized LLMs like the latest open-source DeepSeek model (version V3-0324), you can basically prompt out-of-the-box and create any app and game in -
Google Releases Gemini 2.5 Pro Model to All Users
By
–
Our best model*, Gemini 2.5 Pro, is now available for everyone in the Gemini app model drop down menu at http://
gemini.google.com. Give it a try with your most difficult questions! *For now! -
Gemini App Updates: Canvas, Deep Research, and Free User Rate Limits
By
–
Lots of great new stuff in the Gemini app as of late (Canvas, new models, deep research, personalization, etc), but yes there are rate limits for free users.
-
Use 4o to generate fake documents: 7 examples terrify
By
–
You can use 4o to generate fake documents in seconds. Most verification systems that ask for "just send a photo" are officially obsolete. Here's 7 examples that should terrify everyone: