WarpGBM

⚡ WarpGBM MCP

Neural-speed gradient boosting. GPU-native. MCP-ready.

Outsource your GBDT workload to the world's fastest GPU implementation.
Train on A10G GPUs • Get portable artifacts • Cache for blazing online inference

🎯 What is this? A cloud MCP service that gives AI agents instant access to WarpGBM's GPU-accelerated training. Train models, receive portable artifacts, cache them for millisecond inference. No GPU required on your end.
📚 API Docs 📖 MCP Guide ⭐ GitHub

🚀 How It Works

1. Train: POST your data, get back a portable model artifact
2. Cache: artifact_id is cached for 5 minutes = instant predictions
3. Inference: Online (via cache) or offline (download artifact)

Architecture: Stateless service. No model storage. You own your artifacts. Use them locally, in production, or via our caching layer for fast online serving.

🤖 MCP Integration

Connect AI agents via Model Context Protocol:

  • Endpoint: warpgbm.ai/mcp/sse
  • Tools: train, predict, upload_data, feedback
  • Smart Caching: artifact_id → 5min TTL → sub-100ms inference

View MCP Manifest →

⚡ Model Backends

GPU WarpGBM - 13× faster than LightGBM. Custom CUDA kernels. Invariant learning.
CPU LightGBM - Microsoft's distributed gradient boosting. Battle-tested.

🎯 Iris Dataset Example

Train a multiclass classifier on Iris (60 samples for proper binning):

curl -X POST https://warpgbm.ai/train \
-H "Content-Type: application/json" \
-d '{
"X": [[5.1,3.5,1.4,0.2], [4.9,3,1.4,0.2], [4.7,3.2,1.3,0.2], [4.6,3.1,1.5,0.2], [5,3.6,1.4,0.2],
[7,3.2,4.7,1.4], [6.4,3.2,4.5,1.5], [6.9,3.1,4.9,1.5], [5.5,2.3,4,1.3], [6.5,2.8,4.6,1.5],
[6.3,3.3,6,2.5], [5.8,2.7,5.1,1.9], [7.1,3,5.9,2.1], [6.3,2.9,5.6,1.8], [6.5,3,5.8,2.2],
[7.6,3,6.6,2.1], [4.9,2.5,4.5,1.7], [7.3,2.9,6.3,1.8], [6.7,2.5,5.8,1.8], [7.2,3.6,6.1,2.5]] * 3,
"y": [0,0,0,0,0, 1,1,1,1,1, 2,2,2,2,2,2,2,2,2,2] * 3,
"model_type": "warpgbm",
"objective": "multiclass",
"n_estimators": 100
}'

# Response includes artifact_id for caching:
{
"artifact_id": "abc123...",
"model_artifact_joblib": "H4sIA...",
"training_time_seconds": 0.0
}

# Fast inference with cached artifact:
curl -X POST https://warpgbm.ai/predict_from_artifact \
-d '{"artifact_id": "abc123...", "X": [[5,3.4,1.5,0.2], [6.7,3.1,4.4,1.4], [7.7,3.8,6.7,2.2]]}'

# Predictions: [0, 1, 2] ← Perfect classification!

🐍 Want Local Control? Use the Python Package

For production ML workflows, install WarpGBM directly and use your own GPU:

pip install git+https://github.com/jefferythewind/warpgbm.git

Python Package Benefits:

📦 View on GitHub   |   📘 Agent Guide