Instructions to use kmouratidis/QwenLong-L1-32B-4.25bpw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kmouratidis/QwenLong-L1-32B-4.25bpw with Transformers:

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

pipe = pipeline("text-generation", model="kmouratidis/QwenLong-L1-32B-4.25bpw")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kmouratidis/QwenLong-L1-32B-4.25bpw")
model = AutoModelForCausalLM.from_pretrained("kmouratidis/QwenLong-L1-32B-4.25bpw", 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 kmouratidis/QwenLong-L1-32B-4.25bpw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kmouratidis/QwenLong-L1-32B-4.25bpw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kmouratidis/QwenLong-L1-32B-4.25bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kmouratidis/QwenLong-L1-32B-4.25bpw

SGLang

How to use kmouratidis/QwenLong-L1-32B-4.25bpw 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 "kmouratidis/QwenLong-L1-32B-4.25bpw" \
    --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": "kmouratidis/QwenLong-L1-32B-4.25bpw",
		"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 "kmouratidis/QwenLong-L1-32B-4.25bpw" \
        --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": "kmouratidis/QwenLong-L1-32B-4.25bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kmouratidis/QwenLong-L1-32B-4.25bpw with Docker Model Runner:
```
docker model run hf.co/kmouratidis/QwenLong-L1-32B-4.25bpw
```

kmouratidis

nielsr HF Staff commited on Jun 9, 2025

Commit

7846fe5

verified ·

1 Parent(s): fb95e65

Change library name to transformers (#1)

Browse files

- Change library name to transformers (562c9603f1b029689ce68450e29afebf239a1ce2)

Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>

Files changed (1) hide show

README.md +17 -15

README.md CHANGED Viewed

@@ -1,12 +1,12 @@
 ---
-license: apache-2.0
 base_model:
-  - Tongyi-Zhiwen/QwenLong-L1-32B
-tags:
-  - long-context
-  - large-reasoning-model
 pipeline_tag: text-generation
-library_name: exllamav2
 ---
 ## Quantization
@@ -162,8 +162,10 @@ try:
 except ValueError:
     index = 0
-thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
-content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
 print("thinking content:", thinking_content)
 print("content:", content)
@@ -191,17 +193,17 @@ YaRN is currently supported by several inference frameworks, e.g., `transformers
 - Passing command line arguments:
   For `vllm`, you can use
-  `shell
 vllm serve ... --rope-scaling '{"rope_type":"yarn","factor":4.0,"original_max_position_embeddings":32768}' --max-model-len 131072
-`
   For `sglang`, you can use
-  `shell
 python -m sglang.launch_server ... --json-model-override-args '{"rope_scaling":{"rope_type":"yarn","factor":4.0,"original_max_position_embeddings":32768}}'
-`
   For `llama-server` from `llama.cpp`, you can use
-  `shell
 llama-server ... --rope-scaling yarn --rope-scale 4 --yarn-orig-ctx 32768
-`
   > [!IMPORTANT]
   > If you encounter the following warning
@@ -363,4 +365,4 @@ If you find this work is relevant with your research or applications, please fee
   journal={arXiv preprint arXiv:2505.17667},
   year={2025}
 }
-```

 ---
 base_model:
+- Tongyi-Zhiwen/QwenLong-L1-32B
+library_name: transformers
+license: apache-2.0
 pipeline_tag: text-generation
+tags:
+- long-context
+- large-reasoning-model
 ---
 ## Quantization
 except ValueError:
     index = 0
+thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("
+")
+content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("
+")
 print("thinking content:", thinking_content)
 print("content:", content)
 - Passing command line arguments:
   For `vllm`, you can use
+  ```shell
 vllm serve ... --rope-scaling '{"rope_type":"yarn","factor":4.0,"original_max_position_embeddings":32768}' --max-model-len 131072
+```
   For `sglang`, you can use
+  ```shell
 python -m sglang.launch_server ... --json-model-override-args '{"rope_scaling":{"rope_type":"yarn","factor":4.0,"original_max_position_embeddings":32768}}'
+```
   For `llama-server` from `llama.cpp`, you can use
+  ```shell
 llama-server ... --rope-scaling yarn --rope-scale 4 --yarn-orig-ctx 32768
+```
   > [!IMPORTANT]
   > If you encounter the following warning
   journal={arXiv preprint arXiv:2505.17667},
   year={2025}
 }
+```