Path tracing demo with JS in your web browser. 用浏览器JS做路径跟踪渲染。
仓库
BlinkDL 的仓库
Project Ivory is a simple forum written a few years ago.
The CUDA version of the RWKV language model ( https://github.com/BlinkDL/RWKV-LM )
RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.
RWKV-v2-RNN trained on the Pile. See https://github.com/BlinkDL/RWKV-LM for details.
LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence
My vibe math results
Let us make Psychohistory (as in Asimov) a reality, and accessible to everyone. Useful for LLM grounding and games / fiction / business / finance / governance, and can align agents with human too.
Prepare for DeekSeek R1 inference: Benchmark CPU, DRAM, SSD, iGPU, GPU, ... with efficient code.
An implementation of model parallel autoregressive transformers on GPUs, based on the DeepSpeed library.
LiBai(李白): A Toolbox for Large-Scale Distributed Parallel Training
A *tuned* minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training
RWKV-7: Surpassing GPT
RWKV in nanoGPT style
Understand and test language model architectures on synthetic tasks.