Instructions to use open-machine/Qwen3-1.7B-FlashNorm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use open-machine/Qwen3-1.7B-FlashNorm with Transformers:

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

pipe = pipeline("text-generation", model="open-machine/Qwen3-1.7B-FlashNorm")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("open-machine/Qwen3-1.7B-FlashNorm")
model = AutoModelForCausalLM.from_pretrained("open-machine/Qwen3-1.7B-FlashNorm")
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

vLLM

How to use open-machine/Qwen3-1.7B-FlashNorm with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "open-machine/Qwen3-1.7B-FlashNorm"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "open-machine/Qwen3-1.7B-FlashNorm",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/open-machine/Qwen3-1.7B-FlashNorm

SGLang

How to use open-machine/Qwen3-1.7B-FlashNorm 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 "open-machine/Qwen3-1.7B-FlashNorm" \
    --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": "open-machine/Qwen3-1.7B-FlashNorm",
		"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 "open-machine/Qwen3-1.7B-FlashNorm" \
        --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": "open-machine/Qwen3-1.7B-FlashNorm",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use open-machine/Qwen3-1.7B-FlashNorm with Docker Model Runner:
```
docker model run hf.co/open-machine/Qwen3-1.7B-FlashNorm
```

Qwen3-1.7B-FlashNorm

This is a FlashNorm-prepared checkpoint of Qwen/Qwen3-1.7B, as presented in the paper FlashNorm: Fast Normalization for Transformers by Nils Graef, Matthew Clapp, and Andrew Wasielewski.

Mathematically equivalent to the source model. The per-channel RMSNorm weight tensors (input_layernorm.weight, post_attention_layernorm.weight, model.norm.weight) are folded into the following linear layers and then removed from the state dict entirely.

Framework support note. Stock vLLM currently does not load this checkpoint because the norm weight tensors are absent. The upstream patch to accept missing tensors is tracked at: TBD (vLLM issue link). Until the patch lands, use HuggingFace Transformers; it loads this with a warning that norm weights were not initialized and defaults them to ones, which is the correct behavior for FlashNorm.

What FlashNorm does

An exact reformulation of RMSNorm -> Linear:

Fold the per-channel normalization weight g into the following linear layer: W_star = W @ diag(g), computed once at checkpoint conversion.
After folding, the RMSNorm layer has no learnable per-channel scale. At runtime it simply divides by rms(x).
The resulting model computes the same output as the original, by Proposition 1 of the FlashNorm paper.

See the paper and the transformer-tricks repo for details.

Usage

Regenerate locally with `transformer_tricks`

import transformer_tricks as tt
tt.flashify_repo('Qwen/Qwen3-1.7B', strict=True)

Via HuggingFace Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

tok = AutoTokenizer.from_pretrained('open-machine/Qwen3-1.7B-FlashNorm')
model = AutoModelForCausalLM.from_pretrained('open-machine/Qwen3-1.7B-FlashNorm')

ids = tok('Once upon a time', return_tensors='pt').input_ids
out = model.generate(ids, max_new_tokens=50, do_sample=False)
print(tok.decode(out[0], skip_special_tokens=True))

A warning about missing norm weights is expected; Transformers defaults those to ones, which is the correct value for a FlashNorm checkpoint.

Via vLLM

Not yet supported. See the tracking issue linked above.

License

Inherited from the source model.

Downloads last month: 818

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for open-machine/Qwen3-1.7B-FlashNorm

Base model

Qwen/Qwen3-1.7B-Base

Finetuned

Qwen/Qwen3-1.7B

Finetuned

(721)

this model

Paper for open-machine/Qwen3-1.7B-FlashNorm

Flash normalization: fast RMSNorm for LLMs

Paper • 2407.09577 • Published Jul 12, 2024 • 3