Downloads · 30 days
5
10% of all-time downloads
AJhuggingface/ai-gauge
ai-gauge is a machine learning model from AJhuggingface. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
A fine-tuned Phi-3.5 model for analyzing LLM API calls and recommending cost-effective alternatives.
Downloads · 30 days
5
10% of all-time downloads
All-time downloads
50
Public
Repo size
2.3 GB
Likes
0
Public
Click a slice to open those files.
.gguf2.3 GB · 100%
From the Hugging Face model README
A fine-tuned Phi-3.5 model for analyzing LLM API calls and recommending cost-effective alternatives.
AI-Gauge analyzes your LLM usage patterns and suggests cheaper model alternatives when you're overpaying. It helps developers optimize their AI costs by identifying when expensive models like GPT-5.2 are being used for simple tasks that could be handled by budget models like GPT-4o Mini.
from huggingface_hub import InferenceClient
client = InferenceClient(model="AJhuggingface/ai-gauge", token="your_hf_token")
prompt = """Analyze this LLM API call and determine if the model choice is appropriate:
Model: gpt-5.2
Provider: OpenAI
Input Cost: $1.75/$14.00 per MTok
User Prompt: Fix the typo: 'teh'
Return a JSON object with your analysis..."""
response = client.text_generation(prompt, max_new_tokens=512, temperature=0.1)
# Download and run locally
ollama create ai-gauge -f https://huggingface.co/AJhuggingface/ai-gauge/raw/main/Modelfile
ollama run ai-gauge
The model was fine-tuned on a dataset of real LLM API calls with human annotations for:
AI-Gauge is part of the AI-Gauge VS Code extension, which automatically analyzes your code and provides inline recommendations for cost optimization.
MIT License - see LICENSE file for details.
For questions or contributions, visit the AI-Gauge GitHub repository.