v0.0.21
版本发布时间: 2024-05-11 21:31:56
turboderp/exllamav2最新发布版本:v0.0.21(2024-05-11 21:31:56)
- Support for Granite architecture
- Support for GPT2 architecture
- Support for banned strings in streaming generator
- A bit more work on multimodal support (still unfinished)
- Few bugfixes and stuff
- Windows wheels for PyTorch 2.2.0 are included below to work around an apparent (likely temporary) issue in PyTorch. See #434 and https://github.com/pytorch/pytorch/issues/125109
Full Changelog: https://github.com/turboderp/exllamav2/compare/v0.0.20...v0.0.21
1、 exllamav2-0.0.21+cu117-cp310-cp310-linux_x86_64.whl 95.19MB
2、 exllamav2-0.0.21+cu117-cp310-cp310-win_amd64.whl 95.18MB
3、 exllamav2-0.0.21+cu117-cp311-cp311-linux_x86_64.whl 95.19MB
4、 exllamav2-0.0.21+cu117-cp311-cp311-win_amd64.whl 95.18MB
5、 exllamav2-0.0.21+cu117-cp38-cp38-linux_x86_64.whl 95.19MB
6、 exllamav2-0.0.21+cu117-cp38-cp38-win_amd64.whl 95.18MB
7、 exllamav2-0.0.21+cu117-cp39-cp39-linux_x86_64.whl 95.19MB
8、 exllamav2-0.0.21+cu117-cp39-cp39-win_amd64.whl 95.18MB
9、 exllamav2-0.0.21+cu118-cp310-cp310-linux_x86_64.whl 125.44MB
10、 exllamav2-0.0.21+cu118-cp310-cp310-win_amd64.whl 125.43MB
11、 exllamav2-0.0.21+cu118-cp311-cp311-linux_x86_64.whl 125.45MB
12、 exllamav2-0.0.21+cu118-cp311-cp311-win_amd64.whl 125.43MB
13、 exllamav2-0.0.21+cu118-cp38-cp38-linux_x86_64.whl 125.44MB
14、 exllamav2-0.0.21+cu118-cp38-cp38-win_amd64.whl 125.43MB
15、 exllamav2-0.0.21+cu118-cp39-cp39-linux_x86_64.whl 125.44MB
16、 exllamav2-0.0.21+cu118-cp39-cp39-win_amd64.whl 125.43MB
17、 exllamav2-0.0.21+cu118.torch2.2.0-cp310-cp310-win_amd64.whl 125.43MB
18、 exllamav2-0.0.21+cu118.torch2.2.0-cp311-cp311-win_amd64.whl 125.43MB
19、 exllamav2-0.0.21+cu118.torch2.2.0-cp38-cp38-win_amd64.whl 125.43MB
20、 exllamav2-0.0.21+cu118.torch2.2.0-cp39-cp39-win_amd64.whl 125.43MB
21、 exllamav2-0.0.21+cu121-cp310-cp310-linux_x86_64.whl 126.92MB
22、 exllamav2-0.0.21+cu121-cp310-cp310-win_amd64.whl 126.9MB
23、 exllamav2-0.0.21+cu121-cp311-cp311-linux_x86_64.whl 126.92MB
24、 exllamav2-0.0.21+cu121-cp311-cp311-win_amd64.whl 126.9MB
25、 exllamav2-0.0.21+cu121-cp38-cp38-linux_x86_64.whl 126.92MB
26、 exllamav2-0.0.21+cu121-cp38-cp38-win_amd64.whl 126.9MB
27、 exllamav2-0.0.21+cu121-cp39-cp39-linux_x86_64.whl 126.92MB
28、 exllamav2-0.0.21+cu121-cp39-cp39-win_amd64.whl 126.9MB
29、 exllamav2-0.0.21+cu121.torch2.2.0-cp310-cp310-win_amd64.whl 126.9MB
30、 exllamav2-0.0.21+cu121.torch2.2.0-cp311-cp311-win_amd64.whl 126.9MB
31、 exllamav2-0.0.21+cu121.torch2.2.0-cp38-cp38-win_amd64.whl 126.9MB
32、 exllamav2-0.0.21+cu121.torch2.2.0-cp39-cp39-win_amd64.whl 126.9MB
33、 exllamav2-0.0.21+rocm5.6-cp310-cp310-linux_x86_64.whl 35.14MB
34、 exllamav2-0.0.21+rocm5.6-cp311-cp311-linux_x86_64.whl 35.19MB
35、 exllamav2-0.0.21+rocm6.0-cp310-cp310-linux_x86_64.whl 36.45MB
36、 exllamav2-0.0.21+rocm6.0-cp311-cp311-linux_x86_64.whl 36.52MB
37、 exllamav2-0.0.21-py3-none-any.whl 216.93KB