Spend by model
Priya Sharma, head of platform at Acme Ltd, asked which models drove last month's AI spend. Memtro split cost and requests by model from Portkey's analytics and named the most expensive one.
What are we spending on each model, and which prompt version is live?
Portkey shows you how your AI gateway is used. It reads requests, cost, tokens, latency, errors and cache hit rate over time, split by model, provider, workspace or a metadata key. It also reads gateway configs, prompts and their versions, collections, workspaces and the audit log. When asked and granted, it can record feedback on a trace and create, change, publish or delete a prompt. It can also create, change or delete a config, create a workspace and invite a person. Workspaces, invitations and the audit log need an Admin key. Memtro never reads request logs.
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right Portkey tool, reads what it needs, and answers with the source cited.
These tools change something in Portkey. The model is told to use them only when you ask for exactly that, and an administrator grants each one by name to the teams that may use it; a personal connection acts only after its owner allows it.
Once Portkey is connected, these are the kinds of things people ask. Plain questions, no syntax. If the answer lives partly in another system you have connected, Memtro looks there too.
How much did we spend on gpt-4o last week?
Which model had the highest error rate this month?
What was our cache hit rate over the last 30 days?
Which prompts are in the support collection?
Who changed the production config this week?
Priya Sharma, head of platform at Acme Ltd, asked which models drove last month's AI spend. Memtro split cost and requests by model from Portkey's analytics and named the most expensive one.
Tom Reilly, an engineer at Northwind, asked Memtro to make the latest version of the triage prompt live. With the action granted, Memtro set it as the default, and every caller of the prompt got that version from then on.
The teams that lean on Portkey most, a question each of them asks, and what they connect beside it.
A record in Portkey is rarely the whole story. Connect the systems that hold the rest and Memtro checks them together, then remembers what it learned so the next question starts further along.
Using something else for the same job, or both? Each connector has its own tools; connect as many as you like.
In Portkey open API Keys. Create an Admin API key for the whole organisation (Owner or Admin), or a Workspace API key for one workspace.
In Memtro open Connections → Portkey and paste the API key.
An administrator grants the prompt, config, workspace and invitation actions by name to the teams that may use them. A Workspace key's scopes also limit what it may change.
Ask: how much did we spend on gpt-4o last week?
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The Portkey tools are the same everywhere.
Only when you ask, and only what it has been granted: record feedback (a score) on a request trace; create a prompt template; change a prompt, saving a new version; set which version of a prompt is live; delete a prompt; create a gateway config; change a gateway config that live traffic uses; delete a gateway config; create a workspace; invite a person to the organisation with a role. Every other tool reads, searches or lists. The Connections page labels the tools that act, an administrator grants each one by name to the teams that may use it, and a personal connection acts only after its owner allows it.
A personal connection serves only your own requests. A shared connection serves the organisation: everyone, or only the teams an admin chooses. Admins can always add shared connections; other members need the permission.
Credentials are encrypted at rest. Tool results are cached briefly so agent loops do not repeat identical calls, and the facts the model extracts are saved to your organisation's memory with provenance (system, record and link) so they can be checked and superseded.
API key. In Portkey open API Keys. Create an Admin API key for the whole organisation (Owner or Admin), or a Workspace API key for one workspace.
Seven-day free trial, then one monthly price per organisation. Connectors, jobs and memory are never metered.
Sign in, add a provider key, create an API key and point your tools at Memtro. It takes about five minutes.
7-day free trial · card on file, nothing charged until it ends · cancel any time