RWKV-v2-RNN-Pile
RWKV-CUDA
RWKV-v2-RNN-Pile | RWKV-CUDA | |
---|---|---|
6 | 3 | |
65 | 188 | |
- | - | |
0.0 | 8.5 | |
over 1 year ago | about 2 months ago | |
Python | Cuda | |
Apache License 2.0 | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
RWKV-v2-RNN-Pile
-
[R] RWKV-3: Scaling RNN to 1.5B and Reach Transformer LM Performance (without using attention)
See https://github.com/BlinkDL/RWKV-v2-RNN-Pile for the ppl vs ctxlen curve :)
- [D] Why are transformers still being used?
- [R] RWKV-2 430M release (a parallelizable RNN with transformer-level LM performance, and without using attention)
-
[R] RWKV-v2-RNN : A parallelizable RNN with transformer-level LM performance, and without using attention
Read the inference code in https://github.com/BlinkDL/RWKV-v2-RNN-Pile first :)
RWKV-CUDA
- People who've used RWKV, whats your wishlist for it?
-
Accelerate PyTorch with Taichi: Data Preprocessing & High-performance ML Operator Customization
This repo introduces an interesting example of customizing an ML operator in CUDA. The author developed an RWKV language model using sort of a one-dimensional depthwise convolution custom operator. The model in itself does not involve large amounts of computation, but still runs slow because PyTorch does not have native support for it. So, the author customized the operator in CUDA and used a set of optimization techniques, such as loop fusion and Shared Memory, achieving a performance 20x better than he did with PyTorch.
-
[R] RWKV-v2-RNN : A parallelizable RNN with transformer-level LM performance, and without using attention
It's using my custom CUDA kernel ( https://github.com/BlinkDL/RWKV-CUDA ) to speedup training, so only GPU for now. On the other hand, you don't need CUDA for inference, and it is very fast even on CPUs.
What are some alternatives?
RWKV-LM - 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.
flash-attention - Fast and memory-efficient exact attention
token-shift-gpt - Implementation of Token Shift GPT - An autoregressive model that solely relies on shifting the sequence space for mixing
AI-Writer - AI 写小说,生成玄幻和言情网文等等。中文预训练生成模型。采用我的 RWKV 模型,类似 GPT-2 。AI写作。RWKV for Chinese novel generation.
web-rwkv - Implementation of the RWKV language model in pure WebGPU/Rust.
quality
SmallInitEmb - LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence