BlinkDL ChatRWKV issues

BlinkDL / ChatRWKV

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

Apache License 2.0

9.39k stars 689 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

B

#203 jewc20009 opened 3 weeks ago
0
加速模型载入

#202 shouldsee opened 2 months ago
2
[requires_grad]在本地部署CHATRWKV时遇到了AttributeError: 'str' object has no attribute 'requires_grad'

#201 masterandmiku opened 5 months ago
1
如何选模型基座？

#200 xunyao4dev opened 6 months ago
1
[pip package] feature request: pipeline.generate: add ability to get the state, if it was not provided

#199 Maykeye opened 6 months ago
1
[pip package] Make loading aware that os.environ can change

#198 Maykeye opened 6 months ago
2
add text condition for gen music

#197 jiaqianjing opened 7 months ago
1
model path list

#196 EnricoBeltramo opened 7 months ago
1
mps slower than cpu

#195 fakerybakery opened 7 months ago
1
How to run new v5-Eagle-7B

#194 arun-samespace opened 8 months ago
2
"cpu fp32i8" strategy not working in RWKV v6 through Python rwkv module

#193 polkovnikov closed 7 months ago
2
Inference doesn't work on Apple Macbook even when using CPU fp32 as strategy

#192 fredsco opened 8 months ago
1
eagle-7B

#191 l0d0v1c closed 8 months ago
1
回复总是截断了，如何让回复自然的结束

#189 micronetboy opened 9 months ago
1
feat: add torch parallel mode for v5.2 model.

#188 AsakusaRinne opened 10 months ago
0
大哥，乱码了

#187 humanpp closed 10 months ago
1
NameError: name 'PIPELINE' is not defined

#186 Siu-Ming opened 10 months ago
1
Update rwkv5.cu for multiprocessing

#185 harrisonvanderbyl opened 11 months ago
0
fix bf16

#184 daquexian closed 11 months ago
0
fix multi gpus

#183 daquexian closed 11 months ago
0
fix jit

#182 daquexian closed 11 months ago
0
fix dml sampling

#181 daquexian closed 11 months ago
0
add v5 int8 by unifying float and int8 att

#180 daquexian closed 11 months ago
0
How to write the RWKV in autogressive style like RNN

#179 HaiFengZeng opened 11 months ago
2
remove @MyFunction on cuda_att_seq_v5_2

#178 daquexian closed 12 months ago
0
optimize att_seq_v5_2 by replacing torch.cat with torch.empty

#177 daquexian closed 12 months ago
0
pure pytorch att_seq_v5_2

#176 daquexian closed 12 months ago
0
fix kernel compilation on windows

#175 josStorer closed 12 months ago
0
Support DirectML with minimized code change

#174 pengan1987 closed 12 months ago
0
RuntimeError: Error building extension 'wkv_cuda_v1'

#173 scooter99boston closed 1 year ago
2
[Feature Request] text2music

#172 rayrayraykk opened 1 year ago
2
Prompt for RAG with RWKV-4-World-7B-v1-20230626-ctx4096

#171 Matthieu-Tinycoaching opened 1 year ago
1
fix top_p sampling when cumsum exactly equals to top_p

#170 daquexian closed 1 year ago
0
huggingface无法访问，模型无法下载

#169 xiaokai-lyk closed 1 year ago
4
repalce benchmark.py time with tokens/s

#168 BBuf closed 12 months ago
0
'No CUDA GPUs are available' in google colab with V100 GPU and high RAM

#167 uxff closed 6 months ago
2
demo true error ?

#166 malv-c opened 1 year ago
1
demo ?

#164 malv-c closed 1 year ago
2
开源中文NSFW微调模型

#163 ZhengJun-AI closed 1 year ago
0
fix dense_dim error on pytorch 1.13

#162 daquexian closed 1 year ago
0
add a "stops words" support

#161 yynil opened 1 year ago
0
Add a support to "stop_words" in PIPELINE

#160 yynil opened 1 year ago
0
Feature Request: an option to use Positional Interpolation to extent CTX length at inference time

#159 ytfh44 closed 1 year ago
2
support fp16 on v5 models, fix cublas link errors on some platforms

#158 daquexian closed 1 year ago
0
fast fp16 seq and one mode

#157 daquexian opened 1 year ago
0
fix fp32 run

#156 daquexian closed 1 year ago
0
fix small model fp16 nan

#155 daquexian closed 1 year ago
0
[WIP] Add fast cuda kernels for one mode

#154 daquexian closed 1 year ago
1
求助time-mixing部分，处理数值溢出代码的解释

#153 Ellen7ions closed 1 year ago
0
feat: prefetching

#152 daquexian opened 1 year ago
0