Instructions to use aidonuts/fused-v0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use aidonuts/fused-v0 with Transformers:

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

pipe = pipeline("text-generation", model="aidonuts/fused-v0")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("aidonuts/fused-v0")
model = AutoModelForCausalLM.from_pretrained("aidonuts/fused-v0")
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 aidonuts/fused-v0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "aidonuts/fused-v0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "aidonuts/fused-v0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/aidonuts/fused-v0

SGLang

How to use aidonuts/fused-v0 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 "aidonuts/fused-v0" \
    --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": "aidonuts/fused-v0",
		"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 "aidonuts/fused-v0" \
        --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": "aidonuts/fused-v0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use aidonuts/fused-v0 with Docker Model Runner:
```
docker model run hf.co/aidonuts/fused-v0
```

See axolotl config

axolotl version: 0.6.0

base_model: Qwen/Qwen2.5-Coder-7B-Instruct

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_layer_norm: true
liger_fused_linear_cross_entropy: true

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  # geopandas
  - path: https://www.fused.io/server/v1/realtime-shared/fsh_7UePa8c68x8u89FjmK2Tuu/run/file?dtype_out_vector=parquet
    type: pretrain
    ds_type: parquet
    text_column: text
    split: train
  # examples
  - path: https://staging.fused.io/server/v1/realtime-shared/fsh_2xCVySNfnwmUhWPssX24cn/run/file?dtype_out_raster=png&dtype_out_vector=parquet&cb=12345
    type: pretrain
    ds_type: parquet
    text_column: text
    split: train
  # docs
  - path: https://www.fused.io/server/v1/realtime-shared/fsh_EycsvX70Y3WosxHhdJ8Y9/run/file?dtype_out_raster=png&dtype_out_vector=parquet
    type: pretrain
    ds_type: parquet
    text_column: text
    split: train
  - path: mlabonne/FineTome-100k
    type: chat_template
    split: train[:1%]
    chat_template: qwen_25
    field_messages: conversations
    message_field_role: from
    message_field_content: value

dataset_prepared_path: last_run_prepared
val_set_size: 0.
output_dir: ./outputs/qlora-out

wandb_project: fused-io-copilot
wandb_entity: axolotl-ai
wandb_watch:
wandb_name:
wandb_log_model:

sequence_len: 8192
sample_packing: false
eval_sample_packing: false
pad_to_sequence_len: false


gradient_accumulation_steps: 2
micro_batch_size: 4
num_epochs: 2
optimizer: lion_8bit
lr_scheduler: cosine
learning_rate: 0.00001

train_on_inputs: false
group_by_length: false
bf16: true
fp16:
tf32: true

gradient_checkpointing: true
logging_steps: 1
flash_attention: true

warmup_steps: 20
saves_per_epoch: 1
deepspeed:
weight_decay: 0.01
special_tokens:
  pad_token: "<|end_of_text|>"

save_safetensors: true

outputs/qlora-out

This model is a fine-tuned version of Qwen/Qwen2.5-Coder-7B-Instruct on the https://www.fused.io/server/v1/realtime-shared/fsh_7UePa8c68x8u89FjmK2Tuu/run/file?dtype_out_vector=parquet, the https://staging.fused.io/server/v1/realtime-shared/fsh_2xCVySNfnwmUhWPssX24cn/run/file?dtype_out_raster=png&dtype_out_vector=parquet&cb=12345, the https://www.fused.io/server/v1/realtime-shared/fsh_EycsvX70Y3WosxHhdJ8Y9/run/file?dtype_out_raster=png&dtype_out_vector=parquet and the mlabonne/FineTome-100k datasets.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 8
optimizer: Use OptimizerNames.LION_8BIT and the args are: No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 20
num_epochs: 2