One page for both shapes InferAll serves: ordinary chat completions, and JEV, which answers typed questions about a piece of state and hands back numbers instead of prose. Your key stays in your browser and the request goes straight to the gateway, so this page is the same call you would make from your own code.
Stored in this browser only and sent straight to api.inferall.ai. It never reaches an inferall.ai server, so nothing here can log it. Mint a key. A no-card trial key can call the models marked $0.
The same call as curl
curl https://api.inferall.ai/v1/chat/completions \
-H "Authorization: Bearer ifu_your_key_here" \
-H "Content-Type: application/json" \
-d '{
"model": "meta/llama-3.2-11b-vision-instruct",
"messages": [
{
"role": "user",
"content": "In two sentences, explain what an inference gateway is."
}
],
"max_tokens": 300
}'
Model list verified on 2026-09-24. Every id here was called and checked to answer as itself. Full catalog and pricing.