Instructions to use BeaverAI/Cream-Phi-3-14B-v1a with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BeaverAI/Cream-Phi-3-14B-v1a with Transformers:

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

pipe = pipeline("text-generation", model="BeaverAI/Cream-Phi-3-14B-v1a", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("BeaverAI/Cream-Phi-3-14B-v1a", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("BeaverAI/Cream-Phi-3-14B-v1a", trust_remote_code=True)
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 BeaverAI/Cream-Phi-3-14B-v1a with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BeaverAI/Cream-Phi-3-14B-v1a"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BeaverAI/Cream-Phi-3-14B-v1a",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/BeaverAI/Cream-Phi-3-14B-v1a

SGLang

How to use BeaverAI/Cream-Phi-3-14B-v1a 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 "BeaverAI/Cream-Phi-3-14B-v1a" \
    --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": "BeaverAI/Cream-Phi-3-14B-v1a",
		"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 "BeaverAI/Cream-Phi-3-14B-v1a" \
        --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": "BeaverAI/Cream-Phi-3-14B-v1a",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use BeaverAI/Cream-Phi-3-14B-v1a with Docker Model Runner:
```
docker model run hf.co/BeaverAI/Cream-Phi-3-14B-v1a
```

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

tldr; This is Phi 3 Medium finetuned for (mainly SFW) roleplaying.

It was a promising release candidate that fell flat when things got moist.

I'm publishing all the details for anyone else interested in finetuning Phi 3.

Training Details:

8x H100 80GB SXM GPUs
1 hour training time

Results for Roleplay Mode (i.e., not Instruct format):

Strong RP formatting.
Tends to output short, straightforward replies to the player character.
Starts to break down when things get moist.
Important: My testing is lazy and flawed. Take it with a grain of salt and test the GGUFs before taking notes.

Axolotl Config (some fields omitted)

base_model: failspy/Phi-3-medium-4k-instruct-abliterated-v3
load_in_4bit: true
bf16: auto
fp16:
tf32: false
flash_attention: true

sequence_len: 4096
datasets:
  - path: Undi95/andrijdavid_roleplay-conversation-sharegpt
    type: customphi3

num_epochs: 2
warmup_steps: 30
weight_decay: 0.1

adapter: lora
lora_r: 128
lora_alpha: 16
lora_dropout: 0.1
lora_target_linear: true

gradient_accumulation_steps: 2
micro_batch_size: 2
gradient_checkpointing: true
gradient_checkpointing_kwargs:
   use_reentrant: true

sample_packing: true
pad_to_sequence_len: true

optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 0.0001
max_grad_norm: 1.0

val_set_size: 0.01
evals_per_epoch: 3
eval_max_new_tokens: 128
eval_batch_size: 1

Downloads last month: 845