theluyuan/RWKV-Runner
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
835 Commits 6 Branches 85 Tags 53 MiB
6b4381ee77
Commit Graph

199 Commits

Author SHA1 Message Date
josc146
6b4381ee77 fix #342, #345: cannot import name 'packaging' from 'pkg_resources' 2024-05-28 21:21:45 +08:00
josc146
79476f66a6 deprecate rwkv-beta 2024-05-28 21:15:47 +08:00
josc146
d66fd89947 improve dynamic state api 2024-05-16 13:50:48 +08:00
josc146
b24a18cd3a fix a tps error 2024-05-16 13:48:06 +08:00
josc146
a2bbbabee2 add support for dynamic state-tuned models 2024-05-12 21:51:24 +08:00
josc146
2ddcd17d23 add tps console output 2024-05-10 16:19:21 +08:00
josc146
150bb089cf update state-tuned safetensors converter 2024-05-03 13:10:49 +08:00
josc146
aae7cfe1a2 change the default value of presystem to false 2024-04-30 22:30:06 +08:00
josc146
38b33a7030 upgrade to rwkv 0.8.26 (state instruct align support) 2024-04-30 22:24:22 +08:00
josc146
2d853f92b9 small fix 2024-04-30 21:52:47 +08:00
josc146
16f2201d9f new chat template for /chat/completions (better system support) 2024-03-25 12:52:40 +08:00
josc146
a93610e574 add rwkv version field 2024-03-24 22:29:28 +08:00
josc146
acf5d02104 update global_penalty desc 2024-03-14 12:24:45 +08:00
josc146
bb1a6191b0 prevent 'torch' has no attribute 'cuda' error in torch_gc, so user can use CPU or WebGPU (#302) 2024-03-11 19:04:19 +08:00
josc146
dd89041f72 dep_check.py now ignores GPUtil 2024-03-11 18:55:37 +08:00
dependabot[bot]
f6ba9d7451 Bump fastapi from 0.104.0 to 0.109.1 in /backend-python 
Bumps [fastapi](https://github.com/tiangolo/fastapi) from 0.104.0 to 0.109.1.
- [Release notes](https://github.com/tiangolo/fastapi/releases)
- [Commits](https://github.com/tiangolo/fastapi/compare/0.104.0...0.109.1)

---
updated-dependencies:
- dependency-name: fastapi
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
 2024-03-05 13:51:37 +09:00
dependabot[bot]
96e431e06b Bump python-multipart from 0.0.6 to 0.0.7 in /backend-python 
Bumps [python-multipart](https://github.com/andrew-d/python-multipart) from 0.0.6 to 0.0.7.
- [Release notes](https://github.com/andrew-d/python-multipart/releases)
- [Changelog](https://github.com/Kludex/python-multipart/blob/master/CHANGELOG.md)
- [Commits](https://github.com/andrew-d/python-multipart/compare/0.0.6...0.0.7)

---
updated-dependencies:
- dependency-name: python-multipart
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
 2024-03-05 13:50:47 +09:00
josc146
07d4ba0d6b fix a generation exception caused by potentially dangerous regex being passed into the stop array 2024-03-04 21:20:53 +08:00
josc146
4f14074a75 expose global_penalty 2024-03-02 17:50:41 +08:00
josc146
53a5574080 improve parameters controllable range 2024-03-02 16:52:53 +08:00
josc146
d91c3c004d allow setting tokenChunkSize of WebGPU mode 2024-03-02 16:41:29 +08:00
josc146
887ba06bd6 allow setting quantizedLayers of WebGPU mode; chore 2024-03-01 14:23:05 +08:00
josc146
c9513822c9 fix the issue where state cache could be modified leading to inconsistent hit results 2024-03-01 13:35:16 +08:00
josc146
e3baa0da86 improve occurrence[token] condition 2024-03-01 13:18:03 +08:00
josc146
b0f2ef65d9 improve occurrence[token] condition 2024-02-29 17:54:33 +08:00
josc146
02bbd18acf fix convert_safetensors.py for rwkv6 2024-02-28 23:25:46 +08:00
josc146
18ab8b141f disable AVOID_PENALTY_TOKENS 2024-02-28 23:12:58 +08:00
josc146
771027211a chore 2024-02-22 01:05:52 +09:00
josc146
94fe71b49c change AVOID_PENALTY to \n only 2024-02-22 01:04:05 +09:00
josc146
fafd9f7f6e upgrade to rwkv 0.8.25 2024-02-21 23:50:05 +08:00
Guillermo Marcus
11f1d66383 fix typo in requirements.txt 2024-02-06 19:59:50 +08:00
josc146
3e336830a3 chore 2024-02-06 12:19:12 +08:00
josc146
a1ae71d221 fix /update-config can make the default value of unclearly specified fields invalid by passing in None fields 2024-02-05 22:27:02 +08:00
josc146
19eeeab1e1 add AVOID_PENALTY_TOKENS 2024-02-04 16:49:46 +08:00
josc146
843840baa0 expose penalty_decay, top_k 2024-02-03 22:03:10 +08:00
josc146
26c2954c8e web-rwkv-py 0.1.2 (Support V4, V5 and V6) https://github.com/cryscan/web-rwkv-py 2024-02-03 20:32:23 +08:00
josc146
5f94d86558 add better custom tokenizer support and tokenizer-midipiano.json 2024-02-03 13:04:13 +08:00
josc146
947e127e34 improve path processing 2024-02-02 22:00:01 +08:00
josc146
16b636ef83 add EOS state cache point 2024-01-31 21:33:27 +08:00
josc146
16079a3cba abc music inference support 2024-01-05 12:44:44 +08:00
josc146
b7f4dd835e chore 2023-12-29 00:38:33 +08:00
josc146
7e2380e4ed fix body.state 2023-12-28 23:53:58 +08:00
josc146
7f3cfd54b0 improve state cache performance 2023-12-28 22:15:31 +08:00
josc146
e083f2c629 webgpu(python) state cache 2023-12-28 20:43:57 +08:00
josc146
e33858f110 improve memory usage and speed of convert_safetensors.py 2023-12-26 23:50:51 +08:00
josc146
8ca920a114 release v1.6.6 2023-12-25 21:02:26 +08:00
josc146
a38d5c3a25 enable web-rwkv-py turbo 2023-12-25 20:34:35 +08:00
josc146
a116eff7df webgpu max_buffer_size 2023-12-25 18:08:13 +08:00
josc146
512c4d0f73 improve role-playing effect 2023-12-22 10:51:09 +08:00
josc146
8a3905c09a reduce precompiled web_rwkv_py size 2023-12-15 16:26:01 +08:00