38 1 31

BlinkDL

https://www.rwkv.com/

AI & ML interests

RWKV is all you need

Recent Activity

updated a model 3 days ago

BlinkDL/rwkv-7-world

updated a model 4 days ago

BlinkDL/temp-latest-training-models

updated a Space 4 days ago

BlinkDL/RWKV-Gradio-1

View all activity

Organizations

BlinkDL's activity

updated a model 3 days ago

BlinkDL/rwkv-7-world

Text Generation • Updated 3 days ago • 38

updated a model 4 days ago

BlinkDL/temp-latest-training-models

Updated 4 days ago • 50

updated a Space 4 days ago

Running on T4

464

💻

RWKV-Gradio-1

updated a model 24 days ago

BlinkDL/rwkv-7-pile

Updated 24 days ago • 15

posted an update 24 days ago

Post

2151

RWKV-7 "Goose" 0.4B trained w/ ctx4k automatically extrapolates to ctx32k+, and perfectly solves NIAH ctx16k 🤯 100% RNN and attention-free. Only trained on the Pile. No finetuning. Replicable training runs. tested by our community: https://github.com/Jellyfish042/LongMamba

updated a Space 26 days ago

Running on T4

598

🚀

RWKV-Gradio-2

liked 3 models about 1 month ago

updated a model about 2 months ago

BlinkDL/rwkv-6-misc

Updated Nov 14, 2024 • 19

posted an update about 2 months ago

Post

4669

RWKV-6-world-v3 (+3.1T tokens) is our best multilingual 7B model as of now: BlinkDL/rwkv-6-world

It's 100% RNN and attention-free. MMLU 54.2% (previous world-v2.1 = 47.9%. note: without eval-boosting tricks such as annealing).

RWKV-7-world-v4 soon :)

updated a model about 2 months ago

BlinkDL/rwkv-6-world

Text Generation • Updated Nov 13, 2024 • 142

updated a Space about 2 months ago

Running on T4

464

💻

RWKV-Gradio-1

updated a model 2 months ago

BlinkDL/temp-latest-training-models

Updated 4 days ago • 50

posted an update 4 months ago

Post

5482

RWKV-7 "Goose" preview rc2 => Peak RNN architecture?😃Will try to squeeze more performance for the final release. Preview code & model: https://github.com/BlinkDL/RWKV-LM/tree/main/RWKV-v7