Gateway AI Console

Requests per day

Daily volume (UTC)

Tokens per day

Input + output

Usage by API key

Manage keys under the API Keys tab

Key	Share	Req.	Success	Tokens	Lat. p50	Lat. p95

Usage by runtime backend

Engine model name on the node GPU

Backend	Share	Req.	Success	Tokens	Lat. p50	Lat. p95	TTFB p50

Usage by task

Header X-Gateway-Task on API requests

Task	Share	Req.	Success	Tokens in	Tokens out	Lat. p50

Usage by node route

API model id: node_id/alias (e.g. gpu-01/default)

Route	Share	Req.	Success	Tokens	Lat. p50

API Keys

Create, list, and revoke keys. Usage is under Activity → Usage · keys.

Registered keys

Prefix	Label	Status

Opportunities

Network demand, top models, disk weight (GB), and node runtimes.

Node runtimes (setup)

Docs and install commands. Pick the runtime when registering a model under Nodes.

Top models (7 days)

Gateway telemetry. Disk and VRAM minima are estimates (typical Q4).

Model	Share	Req.	Success	Disk	VRAM min.	Fits GPU?

Demand by alias (7 days)

Alias	Req.	Tokens out	Nodes online	Saturation

Gaps (high demand, low supply)

VRAM and disk guide

Family	Disk	VRAM min.	Fits GPU?	Example pull	Tags

Current public supply

Node route	Runtime	Status	Req. 7d

Nodes

Node health, API clients, GPU load, and registered models. Aggregated usage is under Activity.

Sign in

Activity