Instructions to use SubtleOne/Qwen2.5-32b-Erudite-Writer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SubtleOne/Qwen2.5-32b-Erudite-Writer with Transformers:

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

pipe = pipeline("text-generation", model="SubtleOne/Qwen2.5-32b-Erudite-Writer")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SubtleOne/Qwen2.5-32b-Erudite-Writer")
model = AutoModelForCausalLM.from_pretrained("SubtleOne/Qwen2.5-32b-Erudite-Writer")
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 SubtleOne/Qwen2.5-32b-Erudite-Writer with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SubtleOne/Qwen2.5-32b-Erudite-Writer"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SubtleOne/Qwen2.5-32b-Erudite-Writer",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SubtleOne/Qwen2.5-32b-Erudite-Writer

SGLang

How to use SubtleOne/Qwen2.5-32b-Erudite-Writer 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 "SubtleOne/Qwen2.5-32b-Erudite-Writer" \
    --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": "SubtleOne/Qwen2.5-32b-Erudite-Writer",
		"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 "SubtleOne/Qwen2.5-32b-Erudite-Writer" \
        --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": "SubtleOne/Qwen2.5-32b-Erudite-Writer",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SubtleOne/Qwen2.5-32b-Erudite-Writer with Docker Model Runner:
```
docker model run hf.co/SubtleOne/Qwen2.5-32b-Erudite-Writer
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

This model is a merge using Rombos's top-ranked 32b model, based on Qwen 2.5, and merging three creative writing finetunes. The creative content is a serious upgrade over the base it started with and has a much more literary style than the previous Writer model. I won't call it better or worse, merely a very distinct flavor and style. I quite like it, and enjoin you to try it as well. Enjoy!

GGUF models are avilable here.

merge

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the DELLA merge method using rombodawg/Rombos-LLM-V2.5-Qwen-32b as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

base_model: rombodawg/Rombos-LLM-V2.5-Qwen-32b
parameters:
  int8_mask: true
  rescale: false
  normalize: true
  lambda: 1.04
  epsilon: 0.05
dtype: bfloat16
tokenizer_source: union
merge_method: della
models:
  - model: EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2
    parameters:
      weight: [0.40]
      density: [0.53]
  - model: nbeerbower/Qwen2.5-Gutenberg-Doppel-32B
    parameters:
      weight: [0.30]
      density: [0.53]   
  - model: ArliAI/Qwen2.5-32B-ArliAI-RPMax-v1.3
    parameters:
      weight: [0.40]
      density: [0.53]