A good guide of different local LLM options – and LangChain supports all of them!
OPEN SOURCE
-
Commercializing Superior LLM Models: Safety Certification Strategy
By
–
You've created a superior llama/mistral-derivative model (like @teknium often does).
How can you convince the world to use it (and pay you)? Step 1: You need a 3rd party to approve that this model is safe and responsible.
the Purple Llama project starts to bridge this gap! -
Sakana AI joins Alliance for open AI development approach
By
–
Sakana AI is proud to be a founding member of the AI Alliance, alongside University of Tokyo, Sony and others in Japan. As we’ve seen, the real ‘danger’ of AI is for it to be developed and owned by a single company. An open approach makes much more sense. https://
thealliance.ai -
Open Source Summit: Backend Code Generation Improvements Discussed
By
–
Bye Japan The Open Source Summit was good. Met so many people doing amazing work in the community. Talked with them about how we are making the backend code generation faster. Overall a very nice experience. Sayonara! 🙂
-
Fine-tune LLMs easily with Ludwig AI without expertise
By
–
Can you fine-tune #LLMs without being an expert? Yep, it's possible (hint: open-source @ludwig_ai
) Don't take it from us though. Check out this awesome post from #AI expert and coach, Yogesh Haribhau Kulkarni (PhD). https://
pbase.ai/46Jn3o5 -
Omega: Autonomous LLM Agent for Image Processing Code
By
–
⚡️This is one of the cooler projects we've seen recently
— LangChain (@LangChain) 6 décembre 2023
Omega, an autonomous LLM agent that writes image processing and analysis code, fixes its mistakes, accesses the napari viewer, makes widgets, & more! https://t.co/BrIpRkfxthThis is one of the cooler projects we've seen recently Omega, an autonomous LLM agent that writes image processing and analysis code, fixes its mistakes, accesses the napari viewer, makes widgets, & more!
-
Google adopts Meta’s approach training smaller models massive datasets
By
–
And Google now running with Meta's approach of training smaller models with massive data sets. If you're curious about size, the open source RedPajamas open data set is 30T tokens. We can prob safely assume Google has access to much more.
-
EfficientSAM: Faster Smaller Version of Segment Anything Model
By
–
EfficientSAM !
You liked SAM (Segment Anything Model)?
You're gonna love EfficientSAM!
A smaller, faster, and almost as good version of SAM. -
Meta soutient le mouvement open-source IA
By
–
Kinda surprising that Meta is the one backing the open-source AI movement. Glad to see it happening.