An open question these days is what the role of task-specific / finetuned models will be. I can only think of three scenarios where it makes sense to work on task-specific models. The first scenario is if you have private (e.g., legal, medical, business) data not found on the
LLMS
-
Custom naming for AI model suffixes and full names
By
–
You can use the suffix to custom name part of the model name, but agreed that it would be nice to name the entire thing something custom.
-
LLM Foundation Models: Advances and Applications Explained
By
–
Dive into the world of #LLMs with our series! @matei_zaharia shares what to expect:
— Databricks (@databricks) 19 septembre 2023
✅ The details of LLM foundation models
✅ The latest advances that continue to improve LLMs
✅ Breakthroughs that led to applications like ChatGPT
& more!
Get started ⬇️https://t.co/gw1Pvv3Q92 pic.twitter.com/Xen4ay79JADive into the world of #LLMs with our series! @matei_zaharia shares what to expect: The details of LLM foundation models The latest advances that continue to improve LLMs Breakthroughs that led to applications like ChatGPT
& more! Get started https://
dbricks.co/46g42cX -
OpenAI Launches Fine-Tuning UI with Enhanced Concurrent Training
By
–
The @OpenAI fine-tuning UI is here! You can now see your fine-tunes directly and will be able to create them though the UI in the months to come! We also bumped the concurrent training limit from 1 to 3 so you can fine-tune more models!
-
Hugging Face Releases Open-Access Multimodal AI Model
By
–
Most people missed one of the most important news of the summer – open-access multi-modal (image + text) model coming out of @huggingface
! You can optimize it, fine-tune it and customize it for your use-case: https://
huggingface.co/blog/idefics You can try the fun version of it here: -
SambaNova delivers trillion parameter models on optimized hardware
By
–
“How many resources does it take to run a trillion parameter model like a GPT-4?" With SN40L, @SambaNovaAI can deliver in 8 sockets, on-prem, w' fully optimized hardware delivered. Customers get state of the art accuracy. Coverage via @TechCrunch
-
Building Modern Scalable LLM Platforms: Queensland AI Meetup
By
–
What does it take to build a modern, scalable #LLM platform? Well, if you're in Queensland Australia this week then join Predibase engineering lead Julian Bright at the Queensland AI meetup to learn more. Hope to see you there! Save your spot: https://
pbase.ai/3t1l9kJ -
Groq CEO to Demo LLM Power at SCSP Global Summit
By
–
We're thrilled to be part of this year's amazing line up of speakers at SCSP Global Emerging Technology Summit on Sept 21. Don't miss our CEO & Founder @JonathanRoss321
's demo on the power of #LLMs. @scsp_ai -
SN40L Launch: Hardware Built for Enterprise LLM Applications
By
–
Coverage of the launch of SN40L via @Reuters
. "SN40L is specifically built for large language models running enterprise applications. We've built a full stack that has allowed us to really understand the enterprise use case really well." -
UK Parliament Hearing on AI and Large Language Models
By
–
UK Parliament is holding a hearing on AI & LLMs. Watch it live: https://
parliamentlive.tv/event/index/c6
cee54f-6516-4210-9959-37f75ec6cb71
… Today’s witnesses are Peter Waggett, IBM’s UK Director of Research; BT Group AI Director Zoe Webster; Applied XL Cofounder Francesco Marconi and Air Street Capital Founder Nathan Benaich.