Generate an LLM response
One step with Merge. We can help you get access.
Get something done
Generate an LLM response
with Merge · Model Inference
Use Merge: its Overview Subpackage Responses API lets you generate an LLM response in a single step (POST /responses). capabilities.work can help you get signed up and connected.
- How do I get a model to answer a conversation through the gateway?
- Can I stream an LLM response token by token?
- Is it possible to restrict routing to a specific hosting vendor?
Say it to your AI assistant
- Generate a response to {input} using model {model}.
- Stream a reply to {input} with at most {max_tokens} tokens.
- Answer {input} routed only to vendor {vendor}.
For your developer
POST /responses in Overview Subpackage Responses API. Full spec on APIs.io.