Instructions to use KnutJaegersberg/Tess-M-34B-2bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KnutJaegersberg/Tess-M-34B-2bit with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="KnutJaegersberg/Tess-M-34B-2bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("KnutJaegersberg/Tess-M-34B-2bit")
model = AutoModelForCausalLM.from_pretrained("KnutJaegersberg/Tess-M-34B-2bit", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use KnutJaegersberg/Tess-M-34B-2bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KnutJaegersberg/Tess-M-34B-2bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KnutJaegersberg/Tess-M-34B-2bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/KnutJaegersberg/Tess-M-34B-2bit

SGLang

How to use KnutJaegersberg/Tess-M-34B-2bit with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "KnutJaegersberg/Tess-M-34B-2bit" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KnutJaegersberg/Tess-M-34B-2bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "KnutJaegersberg/Tess-M-34B-2bit" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KnutJaegersberg/Tess-M-34B-2bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use KnutJaegersberg/Tess-M-34B-2bit with Docker Model Runner:
```
docker model run hf.co/KnutJaegersberg/Tess-M-34B-2bit
```

Re-Quantize Model

by igoforth - opened Dec 30, 2023

Discussion

igoforth

Dec 30, 2023

Hi, thank you for your work.

Would you be willing to update the model to support the latest QuiP# changes? I know you opened an issue here https://github.com/Cornell-RelaxML/quip-sharp/issues/31 . You and Minami-su are the only ones I've found who have made QuiP# quantizations so far.

KnutJaegersberg

Owner Dec 30, 2023

It's funny the thought occured to me this morning too: I intuitively assumed I'd have to redo the hessians, which takes ages. Perhaps I only have to redo the latter two steps. That takes less than a day. I'll try that once my GPU is free again.

igoforth

Dec 30, 2023

Thanks! Yeah I had checked that issue a few days ago and the dude mentioned not having to redo the hessians, so that's great news.

Minami-su

Dec 31, 2023

•

edited Dec 31, 2023

It's still available https://github.com/Cornell-RelaxML/quip-sharp/tree/release20231203

KnutJaegersberg

Owner Dec 31, 2023

@igoforth I'm currently quantizing a 70b model, that takes longer than I thought. Maybe that will take 2 more days and I'll be busy with new year, too.

KnutJaegersberg

Owner Dec 31, 2023

80 layers has been running for a day already

KnutJaegersberg

Owner Jan 1, 2024

I finished llama 70b and uploading that now with the newest library version. I'm doing the same now with this 34b model. give it a day or so @igoforth

KnutJaegersberg

Owner Jan 3, 2024

Updated the model.
https://huggingface.co/KnutJaegersberg/Tess-M-34B-2bit

KnutJaegersberg changed discussion status to closed Jan 3, 2024

rakmik

Mar 17, 2025

How do I use the model? I'm tired of kernel build failures.
Is there a way to run inference without building the kernel?

KnutJaegersberg

Owner Mar 18, 2025

I don't think so. better try another quant of another model, this is an old model.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment