-
Before submission delete this line:
List URL:
List Name:
Link to the official homepage of the list manager:
-
## Main Objectives/Goals:
Build CDN with Greenfield
Decentralize the CDN network
Reduce the latency and bandwidth consumption by delivering data from the nearest storage provider
## Challenge Descript…
-
For example, [gemma-2-27b-bnb-4bit](https://huggingface.co/unsloth/gemma-2-27b-bnb-4bit) has 14.6 B parameters, while the main model https://huggingface.co/google/gemma-2-27b, has 27.2 B parameters?
-
### System Info
CPU: Architecture x86_64
GPU: 4xNVIDIA H100
image: [ghcr.io/huggingface/text-generation-inference:sha-a895029](ghcr.io/huggingface/text-generation-inference:sha-a895029)
### Inform…
-
Env: 8 A100-80G GPUs, Transformer: 4.43.3, torch: 2.4.0+cu121
Goal: Get a merged INT8 405B
Path 1:
1) Load IN8 405B & BF16 LoRA --> Merge --> Save
Path 2:
1) Load BF16 405B & BF16 LoRA -->…
-
https://github.com/lllyasviel/stable-diffusion-webui-forge/discussions/981
> Flux Checkpoints
> The currently supported Flux checkpoints are
>
> [flux1-dev-bnb-nf4.safetensors](https://huggingf…
-
```python
from unsloth import FastLanguageModel
model, tokenizer = FastLanguageModel.from_pretrained(
model_name="unsloth/mistral-7b-bnb-4bit",
max_seq_length=2048,
load_in_4bit=Tru…
-
There is a new promising model.
https://huggingface.co/docs/diffusers/main/en/api/pipelines/flux
It would be good to take this model pipeline from diffusers package and add a canvas and attention …
-
Will it be available?
BBCOT updated
1 month ago
-
即使用4090也跑不了meta 8b的模型。感觉是一直在加载模型。我已经把hf上所有文件都下载到llm里。只加载一张图,跑了十几分钟没有变化。
![屏幕截图 2024-09-10 213938](https://github.com/user-attachments/assets/6863875d-a58f-416e-9e09-52b0977aea86)