Instructions to use ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B with Transformers:

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

pipe = pipeline("text-generation", model="ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B")
model = AutoModelForCausalLM.from_pretrained("ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B

SGLang

How to use ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B 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/AHN-GDN-for-Qwen-2.5-Instruct-14B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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/AHN-GDN-for-Qwen-2.5-Instruct-14B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B with Docker Model Runner:
```
docker model run hf.co/ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B
```

Seerkfang commited on Oct 24, 2025

Commit

a4a495c

verified ·

1 Parent(s): 6fa77be

Fix typos in README.md

Browse files

Files changed (1) hide show

README.md +9 -9

README.md CHANGED Viewed

@@ -42,15 +42,15 @@ This repository hosts the model weights for AHN. For installation, usage instruc
 ### Model Zoo
 | base model | AHN module | #params | checkpoint (AHN only) |
 |:---:|:---:| :---:|:---:|
-| Qwen2.5-3B-Instruct | Mamba2 | 119M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-Mamba2-for-Qwen-2.5-Instruct-3B) |
-| Qwen2.5-3B-Instruct | DeltaNet | 118M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-DN-for-Qwen-2.5-Instruct-3B) |
-| Qwen2.5-3B-Instruct | GatedDeltaNet | 130M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-3B) |
-| Qwen2.5-7B-Instruct | Mamba2 | 186M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-Mamba2-for-Qwen-2.5-Instruct-7B) |
-| Qwen2.5-7B-Instruct | DeltaNet | 185M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-DN-for-Qwen-2.5-Instruct-7B) |
-| Qwen2.5-7B-Instruct | GatedDeltaNet | 213M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-7B) |
-| Qwen2.5-14B-Instruct | Mamba2 | 514M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-Mamba2-for-Qwen-2.5-Instruct-14B) |
-| Qwen2.5-14B-Instruct | DeltaNet | 511M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-DN-for-Qwen-2.5-Instruct-14B) |
-| Qwen2.5-14B-Instruct | GatedDeltaNet | 610M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B) |
 ### Evaluation

 ### Model Zoo
 | base model | AHN module | #params | checkpoint (AHN only) |
 |:---:|:---:| :---:|:---:|
+| Qwen2.5-3B-Instruct | Mamba2 | 11.9M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-Mamba2-for-Qwen-2.5-Instruct-3B) |
+| Qwen2.5-3B-Instruct | DeltaNet | 11.8M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-DN-for-Qwen-2.5-Instruct-3B) |
+| Qwen2.5-3B-Instruct | GatedDeltaNet | 13.0M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-3B) |
+| Qwen2.5-7B-Instruct | Mamba2 | 18.6M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-Mamba2-for-Qwen-2.5-Instruct-7B) |
+| Qwen2.5-7B-Instruct | DeltaNet | 18.5M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-DN-for-Qwen-2.5-Instruct-7B) |
+| Qwen2.5-7B-Instruct | GatedDeltaNet | 21.3M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-7B) |
+| Qwen2.5-14B-Instruct | Mamba2 | 51.4M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-Mamba2-for-Qwen-2.5-Instruct-14B) |
+| Qwen2.5-14B-Instruct | DeltaNet | 51.1M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-DN-for-Qwen-2.5-Instruct-14B) |
+| Qwen2.5-14B-Instruct | GatedDeltaNet | 61.0M | [🤗model](https://huggingface.co/ByteDance-Seed/AHN-GDN-for-Qwen-2.5-Instruct-14B) |
 ### Evaluation