https://github.com/blinkdl/chatrwkv

ChatRWKV is like ChatGPT but powered by RWKV (100% RNN) language model, and open source.

Science Score: 46.0%

This score indicates how likely this project is to be science-related based on various indicators:

○
CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
✓
.zenodo.json file
Found .zenodo.json file
○
DOI references
✓
Academic publication links
Links to: arxiv.org
✓
Committers with academic emails
1 of 18 committers (5.6%) from academic institutions
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (10.4%) to scientific vocabulary

Keywords

chatbot chatgpt language-model pytorch rnn rwkv

Keywords from Contributors

transformer diffusers stable-diffusion

Last synced: 5 months ago · JSON representation

Repository

ChatRWKV is like ChatGPT but powered by RWKV (100% RNN) language model, and open source.

Basic Info

Host: GitHub
Owner: BlinkDL
License: apache-2.0
Language: Python
Default Branch: main
Homepage:
Size: 29.8 MB

Statistics

Stars: 9,503
Watchers: 94
Forks: 703
Open Issues: 56
Releases: 0

Topics

chatbot chatgpt language-model pytorch rnn rwkv

Created about 3 years ago · Last pushed 10 months ago

Metadata Files

Readme Funding License

ChatRWKV (pronounced as "RwaKuv" (rʌkuv in IPA), from 4 major params: R W K V)

RWKV homepage: https://www.rwkv.com

RWKV-7 code: https://github.com/BlinkDL/RWKV-LM/tree/main/RWKV-v7

ChatRWKV is like ChatGPT but powered by my RWKV (100% RNN) language model, which is the only RNN (as of now) that can match transformers in quality and scaling, while being faster and saves VRAM.

Our latest version is RWKV-7 https://arxiv.org/abs/2503.14456 (Preview models: https://huggingface.co/BlinkDL/temp )

Gradio Demo 1: https://huggingface.co/spaces/BlinkDL/RWKV-Gradio-1

Gradio Demo 2: https://huggingface.co/spaces/BlinkDL/RWKV-Gradio-2

RWKV-LM main repo: https://github.com/BlinkDL/RWKV-LM (explanation, fine-tuning, training, etc.)

Chat Demo for developers: https://github.com/BlinkDL/ChatRWKV/blob/main/APIDEMOCHAT.py

RWKV Discord: https://discord.gg/bDSBUMeFpc (7k+ members)

Twitter: https://twitter.com/BlinkDL_AI

Homepage: https://www.rwkv.com/

Raw cutting-edge RWKV weights: https://huggingface.co/BlinkDL

HF-compatible RWKV weights: https://huggingface.co/RWKV

Use v2/convert_model.py to convert a model for a strategy, for faster loading & saves CPU RAM.

Note RWKVCUDAON will build a CUDA kernel (much faster & saves VRAM). Here is how to build it ("pip install ninja" first): ```

How to build in Linux: set these and run v2/chat.py

export PATH=/usr/local/cuda/bin:$PATH export LDLIBRARYPATH=/usr/local/cuda/lib64:$LDLIBRARYPATH

How to build in win:

Install VS2022 build tools (https://aka.ms/vs/17/release/vs_BuildTools.exe select Desktop C++). Reinstall CUDA 11.7 (install VC++ extensions). Run v2/chat.py in "x64 native tools command prompt". ``` RWKV pip package: https://pypi.org/project/rwkv/ (please always check for latest version and upgrade)

https://github.com/cgisky1980/ai00rwkvserver Fastest GPU inference API with vulkan (good for nvidia/amd/intel)

https://github.com/cryscan/web-rwkv backend for ai00rwkvserver

https://github.com/saharNooby/rwkv.cpp Fast CPU/cuBLAS/CLBlast inference: int4/int8/fp16/fp32

https://github.com/JL-er/RWKV-PEFT lora/pissa/Qlora/Qpissa/state tuning

https://github.com/RWKV/RWKV-infctx-trainer Infctx trainer

World demo script: https://github.com/BlinkDL/ChatRWKV/blob/main/APIDEMOWORLD.py

Raven Q&A demo script: https://github.com/BlinkDL/ChatRWKV/blob/main/v2/benchmark_more.py

ChatRWKV-strategy

RWKV in 150 lines (model, inference, text generation): https://github.com/BlinkDL/ChatRWKV/blob/main/RWKVin150_lines.py

🔥 RWKV v5 in 250 lines 🔥 (with tokenizer too): https://github.com/BlinkDL/ChatRWKV/blob/main/RWKVv5demo.py

🔥 Building your own RWKV inference engine 🔥: begin with https://github.com/BlinkDL/ChatRWKV/blob/main/src/model_run.py which is easier to understand (used by https://github.com/BlinkDL/ChatRWKV/blob/main/chat.py).

RWKV preprint https://arxiv.org/abs/2305.13048

RWKV-paper

RWKV v6 illustrated:

RWKV-v6

Cool Community RWKV Projects:

https://github.com/saharNooby/rwkv.cpp fast i4 i8 fp16 fp32 CPU inference using ggml

https://github.com/harrisonvanderbyl/rwkv-cpp-cuda fast windows/linux & cuda/rocm/vulkan GPU inference (no need for python & pytorch)

https://github.com/Blealtan/RWKV-LM-LoRA LoRA fine-tuning

https://github.com/josStorer/RWKV-Runner cool GUI

More RWKV projects: https://github.com/search?o=desc&q=rwkv&s=updated&type=Repositories

ChatRWKV v2: with "stream" and "split" strategies, and INT8. 3G VRAM is enough to run RWKV 14B :) https://github.com/BlinkDL/ChatRWKV/tree/main/v2 ```python os.environ["RWKVJITON"] = '1' os.environ["RWKVCUDAON"] = '0' # if '1' then use CUDA kernel for seq mode (much faster) from rwkv.model import RWKV # pip install rwkv model = RWKV(model='/fsx/BlinkDL/HF-MODEL/rwkv-4-pile-1b5/RWKV-4-Pile-1B5-20220903-8040', strategy='cuda fp16')

out, state = model.forward([187, 510, 1563, 310, 247], None) # use 20B_tokenizer.json print(out.detach().cpu().numpy()) # get logits out, state = model.forward([187, 510], None) out, state = model.forward([1563], state) # RNN has state (use deepcopy if you want to clone it) out, state = model.forward([310, 247], state) print(out.detach().cpu().numpy()) # same result as above ``` RWKV-eval

Here is https://huggingface.co/BlinkDL/rwkv-4-raven/blob/main/RWKV-4-Raven-14B-v7-Eng-20230404-ctx4096.pth in action: ChatRWKV

When you build a RWKV chatbot, always check the text corresponding to the state, in order to prevent bugs.

Never call raw forward() directly. Instead, put it in a function that will record the text corresponding to the state.

(For v4-raven models, use Bob/Alice. For v4/v5/v6-world models, use User/Assistant)

The best chat format (check whether your text is of this format): Bob: xxxxxxxxxxxxxxxxxx\n\nAlice: xxxxxxxxxxxxx\n\nBob: xxxxxxxxxxxxxxxx\n\nAlice:

There should not be any space after the final "Alice:". The generation result will have a space in the beginning, and you can simply strip it.
You can use \n in xxxxx, but avoid \n\n. So simply do xxxxx = xxxxx.strip().replace('\r\n','\n').replace('\n\n','\n')

If you are building your own RWKV inference engine, begin with https://github.com/BlinkDL/ChatRWKV/blob/main/src/model_run.py which is easier to understand (used by https://github.com/BlinkDL/ChatRWKV/blob/main/chat.py)

The lastest "Raven"-series Alpaca-style-tuned RWKV 14B & 7B models are very good (almost ChatGPT-like, good at multiround chat too). Download: https://huggingface.co/BlinkDL/rwkv-4-raven

Previous old model results: ChatRWKV

中文模型

QQ群 553456870（加入时请简单自我介绍）。有研发能力的朋友加群 325154699。

中文使用教程：https://zhuanlan.zhihu.com/p/618011122 https://zhuanlan.zhihu.com/p/616351661

推荐UI：https://github.com/l15y/wenda

Star History

Owner

Name: PENG Bo
Login: BlinkDL
Kind: user

Website: https://rwkv.com/
Twitter: BlinkDL_AI
Repositories: 7
Profile: https://github.com/BlinkDL

RWKV is all you need

GitHub Events

Total

Issues event: 5
Watch event: 185
Issue comment event: 3
Push event: 5
Pull request review event: 1
Pull request event: 5
Fork event: 22

Last Year

Issues event: 5
Watch event: 185
Issue comment event: 3
Push event: 5
Pull request review event: 1
Pull request event: 5
Fork event: 22

Committers

Last synced: 9 months ago

All Time

Total Commits: 368
Total Committers: 18
Avg Commits per committer: 20.444
Development Distribution Score (DDS): 0.522

Past Year

Commits: 12
Committers: 3
Avg Commits per committer: 4.0
Development Distribution Score (DDS): 0.333

Top Committers

Name	Email	Commits
BlinkDL	a@a****m	176
PENG Bo	3****L	147
daquexian	d**6@g**m	16
Blealtan Cao	b**n@o**m	8
KerfuffleV2	k**e@k**e	3
cryscan	c**n@u**u	3
oobabooga	1****a	3
EgrorBs	4****s	2
13401015321	k**s@q**m	1
Chui	c**i@c**m	1
MasterYuan418	5****8	1
Pengan	p**7@m**m	1
Zhiyuan Li	u**7@g**m	1
harrisonvanderbyl	3****l	1
josc146	j**r@o**m	1
quantumliu	l**0@s**m	1
tosiyuki	b**n@g**m	1
zk-wz	j**2@1**m	1

Committer Domains (Top 20 + Academic)

126.com: 1 sina.com: 1 msn.com: 1 canva.com: 1 qq.com: 1 umich.edu: 1 keemail.me: 1 a.com: 1

Issues and Pull Requests

Last synced: 6 months ago

All Time

Total issues: 115
Total pull requests: 82
Average time to close issues: 11 days
Average time to close pull requests: 21 days
Total issue authors: 99
Total pull request authors: 43
Average comments per issue: 2.66
Average comments per pull request: 0.66
Merged pull requests: 45
Bot issues: 0
Bot pull requests: 0

Past Year

Issues: 7
Pull requests: 6
Average time to close issues: 4 months
Average time to close pull requests: 7 minutes
Issue authors: 7
Pull request authors: 3
Average comments per issue: 0.29
Average comments per pull request: 0.17
Merged pull requests: 1
Bot issues: 0
Bot pull requests: 0

View more stats

Top Authors

Issue Authors

djaffer (4)
ZTurboX (2)
bello7777 (2)
Maykeye (2)
FreeBlues (2)
fubincom (2)
malv-c (2)
jasonmhead (2)
jcl2023 (2)
lmsdscav (2)
KerfuffleV2 (2)
oobabooga (2)
bennmann (2)
dayu1979 (1)
zhanghui-china (1)

Pull Request Authors

daquexian (15)
BlinkDL (9)
cryscan (6)
Blealtan (4)
zhiyuan1i (3)
EgrorBs (3)
AGENDD (2)
elsatch (2)
Nintorac (2)
hypnopump (2)
kytimmylai (2)
beingPurple (2)
wannaphong (2)
digger-yu (1)
yynil (1)

Top Labels

Issue Labels

Pull Request Labels

Packages

Total packages: 2
Total downloads:
- pypi 8,750 last-month
Total docker downloads: 1,758

Total dependent packages: 1
(may contain duplicates)
Total dependent repositories: 44
(may contain duplicates)
Total versions: 59
Total maintainers: 2

pypi.org: rwkv

The RWKV Language Model

Homepage: https://github.com/BlinkDL/ChatRWKV
Documentation: https://rwkv.readthedocs.io/
License: Apache Software License
Latest release: 0.8.30
published 6 months ago

Versions: 54
Dependent Packages: 1
Dependent Repositories: 44
Downloads: 8,722 Last month
Docker Downloads: 1,758

Rankings

Stargazers count: 0.3%

Forks count: 1.9%

Dependent repos count: 2.2%

Average: 2.7%

Docker downloads count: 3.4%

Downloads: 3.7%

Dependent packages count: 4.8%

Maintainers (1)

BlinkDL

Last synced: 6 months ago

pypi.org: rwkv-beta

The RWKV Language Model

Homepage: https://github.com/BlinkDL/ChatRWKV
Documentation: https://rwkv-beta.readthedocs.io/
License: Apache Software License
Latest release: 0.8.5
published over 2 years ago

Versions: 5
Dependent Packages: 0
Dependent Repositories: 0
Downloads: 28 Last month

Rankings

Dependent packages count: 7.6%

Average: 38.6%

Dependent repos count: 69.5%

Maintainers (1)

daquexian

Last synced: 6 months ago

Dependencies

requirements.txt pypi

prompt_toolkit *
tokenizers >=0.13.2

rwkv_pip_package/pyproject.toml pypi

tokenizers >= 0.13.2

https://github.com/blinkdl/chatrwkv

Science Score: 46.0%

Keywords

Keywords from Contributors

Repository

Basic Info

Statistics

Topics

Metadata Files

README.md

ChatRWKV (pronounced as "RwaKuv" (rʌkuv in IPA), from 4 major params: R W K V)

RWKV homepage: https://www.rwkv.com

RWKV Discord: https://discord.gg/bDSBUMeFpc (7k+ members)

How to build in Linux: set these and run v2/chat.py

How to build in win:

中文模型

Star History

Owner

GitHub Events

Total

Last Year

Committers

All Time

Past Year

Top Committers

Committer Domains (Top 20 + Academic)

Issues and Pull Requests

All Time

Past Year

Top Authors

Issue Authors

Pull Request Authors

Top Labels

Issue Labels

Pull Request Labels

Packages

pypi.org: rwkv

Rankings

Maintainers (1)

pypi.org: rwkv-beta

Rankings

Maintainers (1)

Dependencies