P.S. and its not just limited to LLM finetuning 😀
TOOLS
-
Installation instructions for autotrain-advanced package via pip
By
–
To try, install autotrain-advanced using pip. You can ignore dependencies and install without –no-deps and then you'd need to install some dependencies by hand. "pip install autotrain-advanced" report issues and don't forget to star github repo: https://
github.com/huggingface/au
totrain-advanced
… 2/N -
Hugging Face AutoTrain Client: Fine-tune SOTA Models in Python
By
–
INTRODUCING Hugging Face AutoTrain Client Now you can fine-tune SOTA models on all compatible dataset-model pairs on Hugging Face Hub using Python on Hugging Face Servers. Choose from a number of GPU flavors, millions of models and dataset pairs and 10+ tasks 1/N
-
Triton vs Taichi: PyTorch Integration Framework Comparison
By
–
I went for Triton because it was already closely integrated into PyTorch. Taichi looks really polished in comparison, but not tried it yet!
-
Using GPT-4 for PDF text extraction and summarization
By
–
I've had great success using GPT-4 to extract text and summarize key points from PDFs.
-
Google ReCapture: Generative Video Camera Controls with Masked Fine-Tuning
By
–
Google presents ReCapture
— AK (@_akhaliq) 8 novembre 2024
Generative Video Camera Controls for User-Provided Videos using Masked Video Fine-Tuning pic.twitter.com/5qT386e5K0Google presents ReCapture Generative Camera Controls for User-Provided Videos using Masked Fine-Tuning
-
Mistral Releases Batch API and Moderation API for Cost Efficiency
By
–
Today we are releasing a Batch API and a Moderation API ! The batch API allows to process high-volume requests to all Mistral models at 50% lower cost. https://
mistral.ai/news/batch-api/ https://
mistral.ai/news/mistral-m
oderation/
… -
Mistral Launches Moderation and Batch APIs
By
–
Moderation API – https://
mistral.ai/news/mistral-m
oderation/
…
Batch API – -
Agent Quotas Increased with Usage-Based Billing Expansion
By
–
2/ Unlimited usage with expanded limits: We've INCREASED Agent quotas ($50 / $80 for Core / Teams) and enabled usage-based billing.
— Replit ⠕ (@Replit) 7 novembre 2024
That means even if you hit your limit, you'll always be able to continue using the Agent. pic.twitter.com/MHYQmmkeiw2/ Unlimited usage with expanded limits: We've INCREASED Agent quotas ($50 / $80 for Core / Teams) and enabled usage-based billing. That means even if you hit your limit, you'll always be able to continue using the Agent.
-
Agent Now Supports Multiple Sessions for Projects
By
–
3/ New chats / multi-session support: the Agent now supports multiple sessions for your projects.
— Replit ⠕ (@Replit) 7 novembre 2024
You'll be able to narrow the context of your features, by creating new chats—similar to how you would with ChatGPT or Claude. pic.twitter.com/ARHL0Jy4OJ3/ New chats / multi-session support: the Agent now supports multiple sessions for your projects. You'll be able to narrow the context of your features, by creating new chats—similar to how you would with ChatGPT or Claude.