Instructions to use mindqtrl/qwen3vl-8b-fp8-text-only-en with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mindqtrl/qwen3vl-8b-fp8-text-only-en with Transformers:

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

pipe = pipeline("text-generation", model="mindqtrl/qwen3vl-8b-fp8-text-only-en")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mindqtrl/qwen3vl-8b-fp8-text-only-en")
model = AutoModelForCausalLM.from_pretrained("mindqtrl/qwen3vl-8b-fp8-text-only-en", device_map="auto")
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 mindqtrl/qwen3vl-8b-fp8-text-only-en with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mindqtrl/qwen3vl-8b-fp8-text-only-en"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mindqtrl/qwen3vl-8b-fp8-text-only-en",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mindqtrl/qwen3vl-8b-fp8-text-only-en

SGLang

How to use mindqtrl/qwen3vl-8b-fp8-text-only-en 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 "mindqtrl/qwen3vl-8b-fp8-text-only-en" \
    --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": "mindqtrl/qwen3vl-8b-fp8-text-only-en",
		"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 "mindqtrl/qwen3vl-8b-fp8-text-only-en" \
        --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": "mindqtrl/qwen3vl-8b-fp8-text-only-en",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mindqtrl/qwen3vl-8b-fp8-text-only-en with Docker Model Runner:
```
docker model run hf.co/mindqtrl/qwen3vl-8b-fp8-text-only-en
```

Qwen3-VL 8B FP8 Text-Only English-Only

This is a text-only, English-only variant of the Qwen3-VL 8B FP8 model (ComfyUI mapped, should work with ideogram 4).

Modifications

Vision removed: All model.visual.* tensors (351 tensors) were dropped, leaving only the text decoder (36 layers, 903 tensors).
English-only vocab: Non-English tokens (CJK, Cyrillic, Arabic, etc.) were pruned from the tokenizer and embedding matrix. Vocab reduced from 151,936 to 105,785.
FP8 preserved: The original comfy_quant and weight_scale metadata is intact. No requantization was performed.

Base model

Original: Qwen/Qwen3-VL-8B

Files

model.safetensors — text-only weights (BF16 embeds + FP8 layer weights)
tokenizer.json — pruned BPE tokenizer
config.json — Qwen3ForCausalLM config with updated vocab_size

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("mindqtrl/qwen3vl-8b-fp8-text-only-en")
tokenizer = AutoTokenizer.from_pretrained("mindqtrl/qwen3vl-8b-fp8-text-only-en")

Stats

Metric	Value
Original size	10.59 GB
Text-only size	9.44 GB
English-only size	8.68 GB
Vocab (original)	151,936
Vocab (pruned)	105,785
Layers	36
Hidden size	4096
Attention heads	32
KV heads	8

Downloads last month: 10

Safetensors

Model size

8B params

Tensor type

BF16

F8_E4M3