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

Libraries

How to use ByteDance-Seed/Seed-OSS-36B-Instruct with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ByteDance-Seed/Seed-OSS-36B-Instruct")
model = AutoModelForCausalLM.from_pretrained("ByteDance-Seed/Seed-OSS-36B-Instruct")
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 ByteDance-Seed/Seed-OSS-36B-Instruct with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/ByteDance-Seed/Seed-OSS-36B-Instruct

SGLang

How to use ByteDance-Seed/Seed-OSS-36B-Instruct 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 "ByteDance-Seed/Seed-OSS-36B-Instruct" \
    --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": "ByteDance-Seed/Seed-OSS-36B-Instruct",
		"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 "ByteDance-Seed/Seed-OSS-36B-Instruct" \
        --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": "ByteDance-Seed/Seed-OSS-36B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

vram Requirements for full size

#14

by tazomatalax - opened Aug 24, 2025

Discussion

tazomatalax

Aug 24, 2025

•

edited Aug 24, 2025

Anyone know how much vram we need to run this? Will it run ok on ada 6000 48gb with 100k-ish context?

adamo1139

Aug 24, 2025

Full 36B BF16 model is about 70GB in itself, so it won't fit, but a quant will fit. I am running 4.22bpw EXL3 quant with 150k Q8 context on 2x 3090 Ti 24GB with tensor parallel, it works alright. You can try this, or GGUF quants up to q6_k, sglang/vllm with FP8 , GPTQ-8bit or AWQ/GPTQ 4-bit quants. With 4.22bpw EXL3 and Q4 KV cache you should be able to push up to 300-350k context. I've had normal chats with it till about 120k ctx so far and it was perfectly stable.

Downtown-Case

Aug 25, 2025

•

edited Aug 25, 2025

Anyone know how much vram we need to run this? Will it run ok on ada 6000 48gb with 100k-ish context?

I can fit 100K context in 24GB with exllamav3.

In fact, you'd have room to batch calls in parallel with 48GB if you wish.

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