Repository

Repository di BlinkDL

36 repository supportati

Path tracing demo with JS in your web browser. 用浏览器JS做路径跟踪渲染。

Ultimo commit 11 giu 2018

 (7 star) (2 fork) (0 issue indicizzate) (0 good first issue aperte)

Project Ivory is a simple forum written a few years ago.

Ultimo commit 16 mag 2020

 (0 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)

The CUDA version of the RWKV language model ( https://github.com/BlinkDL/RWKV-LM )

Ultimo commit 10 dic 2025

 (234 star) (35 fork) (0 issue indicizzate) (0 good first issue aperte)

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.

Ultimo commit 14 ago 2026

 (14.665 star) (1019 fork) (0 issue indicizzate) (0 good first issue aperte)

RWKV-v2-RNN trained on the Pile. See https://github.com/BlinkDL/RWKV-LM for details.

Ultimo commit 14 set 2022

 (67 star) (19 fork) (0 issue indicizzate) (0 good first issue aperte)

Ultimo commit 11 ago 2026

 (14 star) (8 fork) (0 issue indicizzate) (0 good first issue aperte)

LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence

Ultimo commit 21 feb 2022

 (61 star) (4 fork) (0 issue indicizzate) (0 good first issue aperte)

My vibe math results

Ultimo commit 16 ago 2026

 (5 star) (2 fork) (0 issue indicizzate) (0 good first issue aperte)

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.

Ultimo commit 9 apr 2023

 (40 star) (1 fork) (0 issue indicizzate) (0 good first issue aperte)

Prepare for DeekSeek R1 inference: Benchmark CPU, DRAM, SSD, iGPU, GPU, ... with efficient code.

Ultimo commit 2 feb 2025

 (73 star) (2 fork) (0 issue indicizzate) (0 good first issue aperte)

An implementation of model parallel autoregressive transformers on GPUs, based on the DeepSpeed library.

Ultimo commit 25 feb 2022

 (0 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)

LiBai(李白): A Toolbox for Large-Scale Distributed Parallel Training

Ultimo commit 5 set 2022

 (0 star) (2 fork) (0 issue indicizzate) (0 good first issue aperte)
BlinkDL/minGPT-tunedJupyter Notebook

A *tuned* minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training

Ultimo commit 9 ago 2021

 (118 star) (11 fork) (0 issue indicizzate) (0 good first issue aperte)

RWKV-7: Surpassing GPT

Ultimo commit 17 nov 2024

 (104 star) (10 fork) (0 issue indicizzate) (0 good first issue aperte)

RWKV in nanoGPT style

Ultimo commit 12 dic 2023

 (200 star) (10 fork) (0 issue indicizzate) (0 good first issue aperte)

Understand and test language model architectures on synthetic tasks.

Ultimo commit 8 giu 2025

 (11 star) (0 fork) (0 issue indicizzate) (0 good first issue aperte)