Video tutorials¶
Each video shows one task from start to finish in the Admin UI and takes less than a minute and a half. Watch them in order if you are new to DeltaLLM. Every video also appears on the guide for its feature, next to the written steps.
All videos are on the DeltaLLM YouTube channel.
1. Add a model and call it¶
Connect a provider model, give it a public name, create an application key, and send a first request with the OpenAI SDK.
Written steps: Find or add your first model and Add and manage models.
2. Route traffic and fail over¶
Put several deployments behind one public name with a route group, send traffic to a backup when a provider fails, and let the auto router pick a model for each request.
Written steps: Route traffic and handle failures and Manage route groups.
3. Set budgets and rate limits¶
Cap spending with organization and team budgets, then assign a tier that sets requests and tokens per minute for each model.
Written steps: Set budgets, Set rate limits, and Set up plans and tiers.
4. Run a batch job¶
Upload a file of inputs, run an embeddings batch for a retrieval (RAG) index, and download the results when the job finishes.
Written steps: Process work in batches and Batch Jobs.