Welcome to TIR AI Platform Documentation!
TIR is a modern AI Development Platform designed to tackle the friction of training and serving large AI models.
We do so by using highly optimised GPU containers (NGC), pre-configured environments (pytorch, tensorflow, triton), automated API generation for model serving, shared notebook storage, and much more.
AI-ML
- Introduction
- Getting Started
- How-to Guides
- Projects
- Notebooks
- Datasets
- Model Repository
- Model Endpoints
- Pipeline
- Run
- Scheduled Run
- Running Docker Image in TIR
- Fine Tuning Models
- WanDb Integration For Fine Tuning
- Guide to Fine Tune Mistral-7B
- Guide to Fine Tune Google Gemma-7b
- Guide to Fine Tune Stable Diffusion XL
- Model Playground
- Samples and Tutorials
- Fine-tune LLMA with Multiple GPUs
- Deploy Inference for Triton
- Deploy Inference for Torchserve
- Deploy Inference for LLMA 2
- Deploy Inference for Codellama
- Deploy Inference for Stable Diffusion v2.1
- Deploy Inference for Stable Video Diffusion XT
- Deploy Inference for Gemma
- Deploy Inference for LLAMA 3 8B-Instruct
- Deploy Inference for YOLOv8
- Fine-tune Stable Diffusion model on TIR
- Deploy Inference for MPT-7B-CHAT
- Custom Containers in TIR
- Fine Tuning Bloom
- Natural Language Queries to SQL with Code Llama
- VLLM with OpenAI client
- Launching LLAMA 3 Inference Using TensorRT-LLM on TIR
- API Tokens
- SSH Key
- Container
- Vector Database
- Mistral7B LLM with RAG for QnA
- Team Features
- Settings
- Analytics
- Billing