Instructions to use frequentor/fahe-v32-sft-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use frequentor/fahe-v32-sft-adapter with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("/home/ubuntu/models/Llama-3.3-70B-Instruct")
model = PeftModel.from_pretrained(base_model, "frequentor/fahe-v32-sft-adapter")

Transformers

How to use frequentor/fahe-v32-sft-adapter with Transformers:

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

pipe = pipeline("text-generation", model="frequentor/fahe-v32-sft-adapter")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("frequentor/fahe-v32-sft-adapter", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use frequentor/fahe-v32-sft-adapter with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "frequentor/fahe-v32-sft-adapter"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "frequentor/fahe-v32-sft-adapter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/frequentor/fahe-v32-sft-adapter

SGLang

How to use frequentor/fahe-v32-sft-adapter 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 "frequentor/fahe-v32-sft-adapter" \
    --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": "frequentor/fahe-v32-sft-adapter",
		"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 "frequentor/fahe-v32-sft-adapter" \
        --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": "frequentor/fahe-v32-sft-adapter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use frequentor/fahe-v32-sft-adapter with Docker Model Runner:
```
docker model run hf.co/frequentor/fahe-v32-sft-adapter
```

FAHE v32.1 — Frequentor AI Hospitality Engine

LoRA adapter (r=64, alpha=128) fine-tuned on Llama-3.3-70B-Instruct for hospitality venue management.

v32.1 is a brevity patch on top of v32: 70 targeted training pairs fixing verbosity failures in structured-response scenarios (str-02, str-03, trd-05). Scored 410/410 on the full FAHE scenario battery.

Training Details

Base: meta-llama/Llama-3.3-70B-Instruct
Source adapter: fahe-v32
Patch pairs: 70 (brevity-focused)
Epochs: 3 | LR: 5e-5 | LoRA r=64, alpha=128
Final loss: 0.035 | Token accuracy: 98.66%

Framework versions

PEFT 0.19.1

Downloads last month: 1

Model tree for frequentor/fahe-v32-sft-adapter

Base model

meta-llama/Llama-3.1-70B

Finetuned

meta-llama/Llama-3.3-70B-Instruct

Adapter

(288)

this model