release v1.6.7
This commit is contained in:
parent
81544ca8b3
commit
94b3882d30
@ -1,12 +1,10 @@
|
|||||||
## Changes
|
## Changes
|
||||||
|
|
||||||
- improve refreshRemoteModels
|
- rwkv5 lora finetune support (https://github.com/JL-er/RWKV-v5-lora)
|
||||||
- reduce precompiled web_rwkv_py size
|
- improve memory usage and speed of convert_safetensors.py
|
||||||
- webgpu(Python) max_buffer_size (12B support) and turbo
|
- webgpu(python) state cache support
|
||||||
- improve role-playing effect
|
- improve state cache performance (especially for rwkv.cpp)
|
||||||
- update manifest.json (a lot of new models)
|
- chore
|
||||||
- bump webgpu(ai00_server) mode to v0.3.8
|
|
||||||
- improve details
|
|
||||||
|
|
||||||
## Install
|
## Install
|
||||||
|
|
||||||
|
Loading…
Reference in New Issue
Block a user