Instructions to use QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4 with Transformers:

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

pipe = pipeline("text-generation", model="QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4")
model = AutoModelForCausalLM.from_pretrained("QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4

SGLang

How to use QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4 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 "QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4 with Docker Model Runner:
```
docker model run hf.co/QuantTrio/Seed-OSS-36B-Instruct-GPTQ-Int4
```

Seed-OSS-36B-Instruct-GPTQ-Int4

File size: 967 Bytes

b3968b9

{
  "name_or_path": "tclf90/Seed-OSS-36B-Instruct-GPTQ-Int4",
  "architectures": [
    "SeedOssForCausalLM"
  ],
  "attention_bias": true,
  "attention_dropout": 0.1,
  "attention_out_bias": false,
  "bos_token_id": 0,
  "pad_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 5120,
  "initializer_range": 0.02,
  "intermediate_size": 27648,
  "max_position_embeddings": 524288,
  "mlp_bias": false,
  "model_type": "seed_oss",
  "num_attention_heads": 80,
  "num_hidden_layers": 64,
  "num_key_value_heads": 8,
  "residual_dropout": 0.1,
  "rms_norm_eps": 1e-06,
  "rope_scaling": {
    "rope_type": "default"
  },
  "rope_theta": 10000000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.55.0",
  "use_cache": true,
  "vocab_size": 155136,
  "quantization_config": {
    "quant_method": "gptq",
    "bits": 4,
    "group_size": 128,
    "sym": false,
    "desc_act": false
  }
}