Instructions to use mombalam/clearledgr-llama-financial-ai with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mombalam/clearledgr-llama-financial-ai with Transformers:

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

pipe = pipeline("text-generation", model="mombalam/clearledgr-llama-financial-ai")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("mombalam/clearledgr-llama-financial-ai", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mombalam/clearledgr-llama-financial-ai with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mombalam/clearledgr-llama-financial-ai"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mombalam/clearledgr-llama-financial-ai",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mombalam/clearledgr-llama-financial-ai

SGLang

How to use mombalam/clearledgr-llama-financial-ai 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 "mombalam/clearledgr-llama-financial-ai" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mombalam/clearledgr-llama-financial-ai",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "mombalam/clearledgr-llama-financial-ai" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mombalam/clearledgr-llama-financial-ai",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mombalam/clearledgr-llama-financial-ai with Docker Model Runner:
```
docker model run hf.co/mombalam/clearledgr-llama-financial-ai
```

Clearledgr Llama Model

Model Description

Clearledgr Llama Model is a specialized financial AI model fine-tuned on Llama 3.1 8B for automated bank reconciliation and financial data processing. This model achieves 87.8% loss reduction and provides high-accuracy financial reconciliation capabilities.

Model Details

Base Model: meta-llama/Llama-3.1-8B
Fine-tuning Method: LoRA (Low-Rank Adaptation)
Training Steps: 9,920
Loss Reduction: 87.8% (from 2.97 to 0.36)
Training Samples: 13,750
LoRA Config: r=8, alpha=32

Capabilities

Bank Reconciliation: 99.7% accuracy
Transaction Matching: Automated matching with AI
Variance Analysis: Precision financial calculations
Exception Handling: Intelligent reconciliation of discrepancies
Multi-format Support: Various bank and GL formats

Training Data

The model was trained on comprehensive financial datasets including:

Bank transaction records
General ledger entries
Reconciliation patterns
Financial compliance rules
International accounting standards

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

# Load base model
base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B")
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.1-8B")

# Load LoRA adapter
model = PeftModel.from_pretrained(base_model, "clearledgr/clearledgr-reconciliation-enhanced-full")

# Generate reconciliation analysis
prompt = "Reconcile the following bank transactions with GL entries..."
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=512)
result = tokenizer.decode(outputs[0], skip_special_tokens=True)

Performance Metrics

Reconciliation Accuracy: 99.7%
Processing Speed: ~2.5 seconds per transaction set
Loss Reduction: 87.8%
Training Efficiency: Converged at 9,920 steps

Deployment

This model is optimized for cloud deployment and supports:

HuggingFace Inference API
Google Colab deployment
RunPod cloud training
Local inference on MacBook M3

License

This model is released under the Llama 3.1 license. See the original Llama 3.1 license for terms and conditions.

Citation

@model{clearledgr2025,
  title={Clearledgr Llama Model},
  author={Clearledgr AI Team},
  year={2025},
  publisher={HuggingFace},
  url={https://huggingface.co/clearledgr/clearledgr-llama-model}
}

Contact

For questions about this model, please contact the Clearledgr team.

Downloads last month: -; Downloads are not tracked for this model. How to track

Model tree for mombalam/clearledgr-llama-financial-ai

Base model

meta-llama/Llama-3.1-8B

Adapter

(736)

this model