Update README.md

README.md

@@ -17,7 +17,12 @@ They're separated by date and commit so it's easier to track any breaking change
 
 If you're here because you want a smaller model to run on a device with constrained memory, try the instruct-based RWKV-Raven ([q8_0](https://huggingface.co/BlinkDL/rwkv-4-raven/tree/main) and [q5_1](https://huggingface.co/latestissue/RWKV-4-Raven-CPP-Converted-Quantized/tree/main)) which goes as low as 1.5B, or [RWKV-PilePlus](https://huggingface.co/BlinkDL/rwkv-4-pileplus/tree/main), which goes as low as 169M.
 
-If you're here because you want an openly-licensed LLaMA, try Together's RedPajama-INCITE, which currently goes [as low as 3B](https://huggingface.co/togethercomputer/RedPajama-INCITE-Base-3B-v1) and [as high as 7B](https://huggingface.co/togethercomputer/RedPajama-INCITE-Base-7B-v0.1). Alternatively, you have MosaicML's MPT, which is [currently only available under 7B](https://huggingface.co/mosaicml/mpt-7b).
+If you're here because you want an openly-licensed LLaMA, there's:
+- OpenLLaMA [(7B)](https://huggingface.co/openlm-research/open_llama_7b_preview_300bt)
+- RedPajama-INCITE [(3B)](https://huggingface.co/togethercomputer/RedPajama-INCITE-Base-3B-v1) [(7B)](https://huggingface.co/togethercomputer/RedPajama-INCITE-Base-7B-v0.1)
+- MPT [(1B)](https://huggingface.co/mosaicml/mpt-1b-redpajama-200b) [(7B)](https://huggingface.co/mosaicml/mpt-7b).
+
+All of them are trained on an open reproduction of LLaMA's dataset, [RedPajama](https://huggingface.co/datasets/togethercomputer/RedPajama-Data-1T), but they're based on different architectures. OpenLLaMA is based on the LLaMA architecture (making it compatible with llama.cpp), RedPajama-INCITE is based on GPT-NeoX, and MPT uses its own. 
 
 # RAM USAGE
 Model | Initial RAM usage