Instructions to use saulpzh/Meta-Task-32B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use saulpzh/Meta-Task-32B with Transformers:

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

pipe = pipeline("text-generation", model="saulpzh/Meta-Task-32B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("saulpzh/Meta-Task-32B")
model = AutoModelForCausalLM.from_pretrained("saulpzh/Meta-Task-32B", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use saulpzh/Meta-Task-32B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/saulpzh/Meta-Task-32B

SGLang

How to use saulpzh/Meta-Task-32B 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 "saulpzh/Meta-Task-32B" \
    --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": "saulpzh/Meta-Task-32B",
		"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 "saulpzh/Meta-Task-32B" \
        --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": "saulpzh/Meta-Task-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use saulpzh/Meta-Task-32B with Docker Model Runner:
```
docker model run hf.co/saulpzh/Meta-Task-32B
```

Meta-Task-32B / README.md

saulpzh

Upload README.md with huggingface_hub

60a1bbc verified 2 days ago

preview code

Raw History Blame Contribute Delete

1.47 kB

metadata

license: apache-2.0
base_model: Qwen/Qwen3-32B
language:
  - en
library_name: transformers
tags:
  - terminal-agents
  - code-agent
  - task-synthesis
  - qwen3

Meta-Task-32B

Qwen3-32B fine-tuned on 3,221 trajectories synthesized by the Meta-Task framework, for solving terminal tasks.

This model accompanies the paper "Meta-Task: Turning Terminal Task Synthesis into a Terminal Task for Scalable Agent Training" (arXiv:2607.27929). Official code: https://github.com/goodmanpzh/meta-task

Results

Terminal-Bench 2.0, Terminus-2 scaffold, non-thinking mode: 31.8 % Avg Pass@1 (base Qwen3-32B: 4.1 %).

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "saulpzh/Meta-Task-32B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="bfloat16", device_map="auto")

Training

LoRA SFT with the SWIFT framework (rank/alpha 64/128, dropout 0.05, lr 1e-4, cosine schedule, 5 epochs, sequence length 40,960, BF16, 8×A100 80GB with DeepSpeed ZeRO-3). Training data: saulpzh/Meta-Task-Data.

License

Apache-2.0.

saulpzh
/

Meta-Task-32B

Meta-Task-32B

Results

Usage

Training

Links

License