Text Generation
Safetensors
English
qwen3
gptqmodel
modelcloud
chat
marin
instruct
int4
gptq
4bit
w4a16
4-bit precision
Instructions to use ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Local Apps Settings
- vLLM
How to use ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16
- SGLang
How to use ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16 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 "ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16" \ --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": "ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16", "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 "ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16" \ --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": "ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16 with Docker Model Runner:
docker model run hf.co/ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16
Download generation_config.json from ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16: direct link, hf CLI and curl.
- Browser
- Download file 234 Bytes
-
https://huggingface.co/ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16/resolve/main/generation_config.json
- Command line
-
hf download hf://ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16/generation_config.json
-
curl -L -o generation_config.json https://huggingface.co/ModelCloud/Marin-32B-Base-GPTQMODEL-W4A16/resolve/main/generation_config.json
234 Bytes
| { | |
| "_from_model_config": true, | |
| "begin_suppress_tokens": [ | |
| 128000, | |
| 128001 | |
| ], | |
| "bos_token_id": 128000, | |
| "decoder_start_token_id": 128000, | |
| "do_sample": true, | |
| "eos_token_id": 128001, | |
| "transformers_version": "4.57.1" | |
| } |