Instructions to use teclabs/llama-capro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use teclabs/llama-capro with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="teclabs/llama-capro",
	filename="llama-capro-f16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use teclabs/llama-capro with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf teclabs/llama-capro:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf teclabs/llama-capro:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf teclabs/llama-capro:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf teclabs/llama-capro:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf teclabs/llama-capro:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf teclabs/llama-capro:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf teclabs/llama-capro:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf teclabs/llama-capro:Q4_K_M

Use Docker

docker model run hf.co/teclabs/llama-capro:Q4_K_M

LM Studio
Jan

vLLM

How to use teclabs/llama-capro with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/teclabs/llama-capro:Q4_K_M

Ollama
How to use teclabs/llama-capro with Ollama:
```
ollama run hf.co/teclabs/llama-capro:Q4_K_M
```

Unsloth Studio new

How to use teclabs/llama-capro with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for teclabs/llama-capro to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for teclabs/llama-capro to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for teclabs/llama-capro to start chatting

Pi new

How to use teclabs/llama-capro with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf teclabs/llama-capro:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "teclabs/llama-capro:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use teclabs/llama-capro with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf teclabs/llama-capro:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default teclabs/llama-capro:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use teclabs/llama-capro with Docker Model Runner:
```
docker model run hf.co/teclabs/llama-capro:Q4_K_M
```

Lemonade

How to use teclabs/llama-capro with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull teclabs/llama-capro:Q4_K_M

Run and chat with the model

lemonade run user.llama-capro-Q4_K_M

List all available models

lemonade list

llama-capro / README.md

teclabs

Upload README.md with huggingface_hub

678e29e verified 6 months ago

preview code

raw

history blame contribute delete

2.83 kB

metadata

license: llama3.1
base_model: meta-llama/Llama-3.1-8B-Instruct
tags:
  - llama
  - llama-3
  - gguf
  - quantized
  - indian-accounting
  - ind-as
  - finance
  - accounting
  - ca
language:
  - en
pipeline_tag: text-generation

Llama-CAPRO (CA Professional) - Indian Accounting Standards Expert

CAPRO is a specialized AI assistant for Chartered Accountants and finance professionals, fine-tuned on Indian Accounting Standards (Ind AS).

Model Details

Base Model: meta-llama/Llama-3.1-8B-Instruct
Fine-tuned Adapter: 0xadityam/llama-aica
Domain: Indian Accounting Standards (Ind AS)
Format: GGUF (Ollama, LM Studio, llama.cpp ready)
Training: LoRA fine-tuned on 269 Ind AS examples

Available Quantizations

Format	Size	Use Case	RAM Required
F16	~15GB	Maximum quality	20GB+
Q5_K_M	~5.5GB	Good quality, laptops	8-12GB
Q4_K_M	~4.5GB	Balanced, edge devices	6-8GB

Quick Start with Ollama

Download Q4_K_M (recommended) wget https://huggingface.co/teclabs/llama-capro/resolve/main/llama-capro-q4_k_m.gguf

Create model ollama create llama-capro -f Modelfile-q4-k-m

Run ollama run llama-capro

Sample Queries

"What is the objective of Ind AS 1?"
"Explain revenue recognition under Ind AS 115"
"What are disclosure requirements for financial instruments?"
"How should goodwill be accounted for under Ind AS?"
"What is the difference between Ind AS and IFRS?"

Modelfile (Ollama)

FROM llama-capro-q4_k_m.gguf

PARAMETER temperature 0.7 PARAMETER top_p 0.9 PARAMETER stop "<|eot_id|>" PARAMETER num_ctx 2048

SYSTEM "You are CAPRO (CA Professional), an expert on Indian Accounting Standards (Ind AS). Provide accurate answers about Ind AS regulations and accounting policies."

Use with LM Studio

Download LM Studio from lmstudio.ai
Click "Import" → "Import GGUF"
Select the downloaded .gguf file
Start chatting

Training Details

LoRA Rank: 64
LoRA Alpha: 16
Training Epochs: 3
Learning Rate: 2e-4
Max Sequence Length: 2048
Training Data: 269 examples covering Ind AS 1 and related standards

Limitations

Specialized for Indian Accounting Standards
Not suitable for other accounting standards (IFRS, US GAAP)
Responses should be verified with official documents
Not for legal or investment advice

License

Llama 3.1 Community License

Acknowledgements

Base model: meta-llama/Llama-3.1-8B-Instruct
LoRA adapter: 0xadityam/llama-aica
Conversion: llama.cpp

CAPRO - CA Professional AI Assistant
Generated: 2025-11-14