Instructions to use cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated with Transformers:

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

pipe = pipeline("image-text-to-text", model="cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated")
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)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated")
model = AutoModelForMultimodalLM.from_pretrained("cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated

SGLang

How to use cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated 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 "cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated" \
    --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": "cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated" \
        --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": "cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated with Docker Model Runner:
```
docker model run hf.co/cbert33/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated
```

Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-Calibrated / RELEASE_MANIFEST.json

cbert33

Upload model card and metadata

7a0149c verified 2 days ago

Raw

History Blame Contribute Delete

2.51 kB

	{
	"schema_version": 1,
	"created_at": "2026-09-13T05:58:52.512198+00:00",
	"model": "Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-FP8-KV-Calibrated",
	"source": {
	"repo_id": "DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored",
	"revision": "9cdc928b322203a58c362584882105a56918a640"
	},
	"artifact": {
	"architecture": "Qwen3_5ForConditionalGeneration",
	"format": "safetensors/compressed-tensors",
	"quantization": "FP8_BLOCK weights, dynamic FP8 activations, calibrated static tensor-wise FP8 KV scales",
	"files": 22,
	"bytes": 31266097056
	},
	"templates": {
	"default": "chat_template.jinja",
	"davidau_original": "chat_template-davidau-original.jinja",
	"qwen_original": "chat_template-qwen-original.jinja"
	},
	"validation": "VALIDATION.json",
	"checksums": "SHA256SUMS",
	"files": [
	{
	"path": "README.md",
	"bytes": 4514
	},
	{
	"path": "README.upstream.md",
	"bytes": 9852
	},
	{
	"path": "VALIDATION.json",
	"bytes": 2123
	},
	{
	"path": "chat_template-davidau-original.jinja",
	"bytes": 16835
	},
	{
	"path": "chat_template-old.jinja",
	"bytes": 17180
	},
	{
	"path": "chat_template-qwen-original.jinja",
	"bytes": 8952
	},
	{
	"path": "chat_template-toolcall.jinja",
	"bytes": 16835
	},
	{
	"path": "chat_template-tturbo.jinja",
	"bytes": 38789
	},
	{
	"path": "chat_template.jinja",
	"bytes": 16835
	},
	{
	"path": "config.json",
	"bytes": 21542
	},
	{
	"path": "generation_config.json",
	"bytes": 202
	},
	{
	"path": "model-00001-of-00002.safetensors",
	"bytes": 19952396812
	},
	{
	"path": "model-00002-of-00002.safetensors",
	"bytes": 10437273492
	},
	{
	"path": "model.safetensors.index.json",
	"bytes": 156590
	},
	{
	"path": "model_mtp.safetensors",
	"bytes": 849400392
	},
	{
	"path": "preprocessor_config.json",
	"bytes": 390
	},
	{
	"path": "processor_config.json",
	"bytes": 1191
	},
	{
	"path": "recipe.yaml",
	"bytes": 896
	},
	{
	"path": "tokenizer.json",
	"bytes": 19989325
	},
	{
	"path": "tokenizer_config.json",
	"bytes": 1165
	},
	{
	"path": "video_preprocessor_config.json",
	"bytes": 385
	},
	{
	"path": "vocab.json",
	"bytes": 6722759
	}
	]
	}