Ollama Cloud
Run prompts, chats and embeddings against Ollama's hosted cloud API with an API key.
About
Ollama Cloud ships in the w6w first-party pack. It declares 5 actions, 3 health checks, and the host runs its code in a sandbox that never sees the credential.
- App id
io.w6w.ollama-cloud- Version
- 0.1.0
- Author
- w6w
- Licence
- MIT
- Categories
- AI & Machine Learning
Overview
Ollama Cloud serves the same models as a local Ollama install, hosted by the vendor, and this app calls it directly: you paste an API key from ollama.com and the app talks to https://ollama.com with it. A workflow can generate text from a prompt, continue a chat conversation, produce embeddings, and ask for structured JSON that matches a schema, with images for vision models and sampling options such as temperature, seed and context size. The models the account can call can be listed, and the service version read, so a workflow’s model choice stays explicit rather than hardcoded.
The difference from the Ollama app is where the model runs. There is no server to host, no LAN address to reach and no hardware to size: the trade is an API key, a usage limit and the vendor’s model list instead of your own. Every call is non-streaming. Model management stays with the vendor — the cloud API exposes no pull, copy, delete or unload — so a model the account does not have access to answers 404, and List models shows what it does.
Build with Ollama Cloud
Three routes to the same 5 actions. The Workflow tab is generated from Ollama Cloud's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
Chat with a model
chat Send a conversation (a messages array) to a model on Ollama Cloud and get the assistant's reply as one result (non-streaming). Supports images, tools, JSON or JSON-Schema structured output, and sampling options. Ollama keeps no history: send all of it every time.
Create embeddings
embed Turn text into embedding vectors with an embedding model on Ollama Cloud. Takes one string or a JSON array of strings and always returns a list of vectors.
Generate a completion
generate Send a prompt to a model on Ollama Cloud and get the full response back as one result (non-streaming). Supports a system prompt, images for vision models, JSON or JSON-Schema structured output, and sampling options.
List installed models
model-list The models this Ollama Cloud account can call, with size, digest, family, parameter size, quantization and capabilities.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying Ollama Cloud's real ids.
{
"manifestVersion": "2",
"name": "ollama-cloud-example",
"steps": [
{
"id": "chat",
"uses": {
"app": "io.w6w.ollama-cloud",
"action": "chat",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"model": "<model>",
"messages": "<messages>"
}
}
]
}Here are some of the things you can do
-
Chat with a model
performchat -
Create embeddings
performembed -
Generate a completion
performgenerate -
List installed models
readmodel-list -
Get service version
readversion
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for Ollama Cloud, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to Ollama Cloud. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "chat",
payload: {
model: "<value>",
messages: "<value>",
// format: "<value>",
// tools: "<value>",
// think: "<value>",
// temperature: "<value>",
// seed: "<value>",
// numPredict: "<value>",
// numCtx: "<value>",
// options: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value);npm install -g @w6w/cliw6w run conn_YOUR_CONNECTION_ID --action chat --payload '{"model":"<value>","messages":"<value>"}'Give an AI agent Ollama Cloud — without giving it Ollama Cloud's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.ollama-cloud#chat",
"input": {
"model": "<model>",
"messages": "<messages>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's Ollama Cloud connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
What the agent gets
Credentials it can't read
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
A tool surface scoped to the caller
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
A durable workflow in one call
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Health-aware discovery
Ollama Cloud's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. Ollama Cloud itself is MIT, and the runtime that executes it is source-available (FSL).
Health checks
Ollama Cloud declares its own checks, so its health is a property of the app rather than something the host guesses at.
Ollama Cloud platform status
Ollama Cloud reachable
Whether Ollama Cloud answers GET /api/version with this connection's API key.
API request headroom
Requests left in the current Ollama Cloud limit window, read off the response headers when the API sends them.