Instructions to use FINAL-Bench/Darwin-397B-ZTC with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FINAL-Bench/Darwin-397B-ZTC with Transformers:

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

pipe = pipeline("text-generation", model="FINAL-Bench/Darwin-397B-ZTC")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("FINAL-Bench/Darwin-397B-ZTC")
model = AutoModelForMultimodalLM.from_pretrained("FINAL-Bench/Darwin-397B-ZTC", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use FINAL-Bench/Darwin-397B-ZTC with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "FINAL-Bench/Darwin-397B-ZTC"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "FINAL-Bench/Darwin-397B-ZTC",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/FINAL-Bench/Darwin-397B-ZTC

SGLang

How to use FINAL-Bench/Darwin-397B-ZTC 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 "FINAL-Bench/Darwin-397B-ZTC" \
    --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": "FINAL-Bench/Darwin-397B-ZTC",
		"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 "FINAL-Bench/Darwin-397B-ZTC" \
        --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": "FINAL-Bench/Darwin-397B-ZTC",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use FINAL-Bench/Darwin-397B-ZTC with Docker Model Runner:
```
docker model run hf.co/FINAL-Bench/Darwin-397B-ZTC
```

Darwin-397B-ZTC / README.md

Commit History

Sep 28 23 days ago

03:29
Link successor Darwin-180B-RSI (#1 GPQA Diamond)

SeaWolf-AI

Verified signature
33a193f

Sep 21 30 days ago

23:25
Drop derivative-signal tags (quantized/fp8/w8a8/compressed-tensors); FP8 remains stated in the card body

SeaWolf-AI

Verified signature
8701335
06:42
정정: 폐기값(0.7364/0.7282) → 정본(0.7272/0.7289) + 정정 고지 추가

SeaWolf-AI

Verified signature
7ff4f83

Sep 20

23:22
card: limitation — score shifts when the answering-model mixture changes (measured)

SeaWolf-AI

Verified signature
30cd2de
22:55
card: generate candidates and let the verifier pick (+2.31 pp); majority voting measured as a loss

SeaWolf-AI

Verified signature
4336cc1
22:52
card: JEV-compatible API, calibration, and what to do on a low score

SeaWolf-AI

Verified signature
a5fa3d0
16:41
card: independent leaderboard 0.7364, per-domain vs baseline, protocol separation

SeaWolf-AI

Verified signature
7242852
12:20
docs: cross-link ZTC ecosystem badges

SeaWolf-AI

Verified signature
8e19303
12:12
docs: cross-link ZTC ecosystem badges

SeaWolf-AI

Verified signature
868fd5b
03:27
card: explicit lineage table crediting Qwen/Qwen3.5-397B-A17B

SeaWolf-AI

Verified signature
6f10591
03:27
card: attribution moved to body (matches Darwin family convention); base model credited in lineage table and citation

SeaWolf-AI

Verified signature
886b0ab
03:25
card: drop base_model_relation (this is a Darwin V9 model, not a quantization of the base)

SeaWolf-AI

Verified signature
dbd92b9
03:23
card: document the shipped ZTC probe

SeaWolf-AI

Verified signature
4c84f05
03:21
Darwin-397B-ZTC: FP8 397B MoE with Zero-Token Confidence (GPQA Diamond 93.43)

SeaWolf-AI

Verified signature
91c58e3

Jun 25

08:00
Add files using upload-large-folder tool

SeaWolf-AI

Verified signature
057fbce

Commit History

Sep 28 23 days ago

Link successor Darwin-180B-RSI (#1 GPQA Diamond)

Sep 21 30 days ago

Drop derivative-signal tags (quantized/fp8/w8a8/compressed-tensors); FP8 remains stated in the card body

정정: 폐기값(0.7364/0.7282) → 정본(0.7272/0.7289) + 정정 고지 추가

Sep 20

card: limitation — score shifts when the answering-model mixture changes (measured)

card: generate candidates and let the verifier pick (+2.31 pp); majority voting measured as a loss

card: JEV-compatible API, calibration, and what to do on a low score

card: independent leaderboard 0.7364, per-domain vs baseline, protocol separation

docs: cross-link ZTC ecosystem badges

docs: cross-link ZTC ecosystem badges

card: explicit lineage table crediting Qwen/Qwen3.5-397B-A17B

card: attribution moved to body (matches Darwin family convention); base model credited in lineage table and citation

card: drop base_model_relation (this is a Darwin V9 model, not a quantization of the base)

card: document the shipped ZTC probe

Darwin-397B-ZTC: FP8 397B MoE with Zero-Token Confidence (GPQA Diamond 93.43)

Jun 25

Add files using upload-large-folder tool