-
### Your current environment
Hey Guys,
I tried the open ai api server, to load a 70B Llama-3 checkpoint. I think out of the 3-4 efforts I did, only one time the model successfully loaded after a…
-
### What is the issue?
I downloaded the codegemma and codellama models from Huggingface and fine tuned them using llama factory. After importing the fine tuned model into Ollama, Codellama works norm…
-
### Description of defect
In our mbed_app.json apply the following configuration:
```json
"target_overrides": {
"*": {
....
"lora.phy": "AS923",
…
-
使用的是秋叶整合包，直接搜索报错说让将torch回退到21coda118，我也不是很懂，直接在整合包里回退了，
之前还有别的状况，我在custom_nodes下按照主页里的方法用指令下载依赖后出现闪退，我按照其他教程…
-
### Your current environment
- vLLM CPU : v0.6.0
- Hardware: Intel(R) Xeon(R) Platinum 8480+ CPU
- Model: google/gemma-2-2b
### 🐛 Describe the bug
vLLM v0.6.0 (cpu) is throwing below erro…
-
!swift sft \
--model_type got-ocr2 \
--model_id_or_path stepfun-ai/GOT-OCR2_0 \
--sft_type lora \
--dataset /kaggle/working/output_data.json \
--output_dir /kaggle/working/hindi_got_model_3 \
--…
-
When I try to use `compile_model` with CUDA as the specified device, I encounter the following error. Is there a way to resolve this, or is the `lora.py` code not yet compatible with running on a GPU?…
-
### What happened?
there seems to be an issue with the rescale noise scheduler option on v-pred ztsnr sdxl models
all 12 tries resulted in either no change at all to the model or the outputs slowly …
-
### Is this a bug or a feature request?
bug
### What did you expect?
build lora-app-server on Raspberry pi 3 , running alpine arm64
### What happened?
```
Generating combined Swagger JSON
[…
-
### 🚀 The feature, motivation and pitch
I am a developer from Cambricon, an AI chip vendor in China. We have already supported vLLM 0.6.1.post2 on Cambricon MLU internally. We wish to contribute th…