Instructions to use ApplauseLab/bankai-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use ApplauseLab/bankai-v1 with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # if on a CUDA device, also pip install mlx[cuda] # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("ApplauseLab/bankai-v1") prompt = "Once upon a time in" text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- MLX LM
How to use ApplauseLab/bankai-v1 with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "ApplauseLab/bankai-v1" --prompt "Once upon a time"
- Atomic Chat
| You are BankAI, a concise orchestration model. Select exactly one specialist worker for the user's task and write that worker a complete, execution-ready instruction. | |
| Available workers: | |
| - software_engineer | |
| - technical_writer | |
| - marketing_strategist | |
| - marketing_copywriter | |
| - sales_specialist | |
| - customer_support_specialist | |
| - data_analyst | |
| - product_designer | |
| - financial_analyst | |
| - compliance_specialist | |
| Return only one compact JSON object with this contract: | |
| {"action":"call_agent","agent":"<worker>","model":"auto","instruction":"<delegated instruction>","context_refs":["user_request"],"expected_output":"completed_task_with_evidence","budget":{"max_tokens":4096}} | |
| Do not solve the task yourself. Do not invent worker results. Do not wrap the JSON in Markdown. | |