RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
Apache License 2.0
12.32k
stars
838
forks
source link
Add LocalAI in the Community projects README.md #102
Hey :wave: this is an amazing project - thanks @BlinkDL and @saharNooby for rwkv.cpp!
I've assembled the bits to add support to rwkv in https://github.com/go-skynet/LocalAI (https://github.com/go-skynet/LocalAI/pull/158), so now it allows rwkv-ggml models to be used with a RESTful API compatible with OpenAI clients, so I thought it could be a good addition to the community projects as it is a another concrete use-case!
I thought could be helpful for other people out in the community here trying to add rwkv into their applications, so I've opened up this PR to add LocalAI to the community projects. Feel free to close it if you don't think it fits!
Hey :wave: this is an amazing project - thanks @BlinkDL and @saharNooby for rwkv.cpp!
I've assembled the bits to add support to rwkv in https://github.com/go-skynet/LocalAI (https://github.com/go-skynet/LocalAI/pull/158), so now it allows rwkv-ggml models to be used with a RESTful API compatible with OpenAI clients, so I thought it could be a good addition to the community projects as it is a another concrete use-case!
I thought could be helpful for other people out in the community here trying to add rwkv into their applications, so I've opened up this PR to add LocalAI to the community projects. Feel free to close it if you don't think it fits!