Have you ever wanted to train LLMs in pure C without 245MB of PyTorch and 107MB of cPython? No? Well now you can! With llm.c:
To start, implements GPT-2 training on CPU/fp32 in only ~1,000 lines of clean code. It compiles and runs instantly, and exactly
I'm extremely excited to announce "the big bomb"!: Neo and Matrix, that we're working on with colleagues and friends from open-source community, , wuhan ai, and . Neo is the first fully-transparent bilingual large language model, with
" LLAMA3 still suffers non-negligent degradation in these scenarios, especially in ultra-low bit-width. "
Very interesting paper in the Large Language Model space named "How Good Are Low-bit Quantized LLAMA3 Models? An Empirical Study"
📌 This research dives deep into
Want to see a real-world RAG app in production? Check out wandbot 🤖 - chat over
@weights_biases
documentation, integrated with
@Discord
and
@SlackHQ
! (Full credits to
@ParamBharat
et al.)
It contains the following key features that every user should consider:
✅ Periodic data
Which mix of embeddings/rerankers works best for RAG?
Thanks to
@ravithejads
, we’ve created our most comprehensive survey yet, analyzing 7 SOTA embedding models and 3 rerankers 🔥.
Best of all, the full Colab notebook is provided so you can 1) auto-generate an eval dataset, and