Guide to TitanML...

These docs are outdated! Please check out https://docs.titanml.co for the latest information on the TitanML platform. If there's anything that's not covered there, please contact us on our discord.

Titan Takeoff 🛫

Titan Takeoff 🛫: Inference Server

What does it do?

• Super fast optimised inference, even on local hardware like CPUs

• Quickly experiment with inferencing different LLMs

• Create local versions of ChatGPT & The Playground

• Create inference servers that are local and private (think HF Inference Servers but local)

• All for Generative models

Supported models: Most OS Generative model architectures

Titan Train 🎓

Titan Train 🎓: Finetuning Service

What does it do?

• Fine-tuning of language models

• Using QLoRA for super efficient training

• Super simple, only a few lines of code

• Don't worry about infrastructure - all hosted by TitanML

Supported models: Both generative and non-generative language models

Titan Optimise ✨

Titan Optimise ✨: Knowledge Distillation

What does it do?

• Compression of Natural Language Understanding tasks

• Helps when latency, memory, or cost is a severe bottleneck

• Uses the latest compression techniques like pruning & knowledge distillation for non-generative tasks

Supported models: Non-generative models