Dust vs Gemini Enterprise

Dust

The whole platform is MIT on GitHub, with a public OpenAPI document of 68 operations, llms.txt and a remote MCP server that acts with the signed-in user's access. API calls that run agents draw only on a paid workspace credit pool, and the terms of service could not be read.

Gemini Enterprise

The assistant is callable through a GA v1 method with a 99.5 per cent SLA, OAuth through IAM, a public Discovery document and optional Model Armor screening. Seat prices weren't found on Google's pages, onboarding needs a billing account and a person, and some Gemini agent capabilities announced on 8 October are early access.

Dust

1. Use https://eu.dust.tt for a workspace in the EU region and https://dust.tt otherwise, for both the API and the MCP server
2. Connect an MCP client by OAuth and send the resource parameter in authorisation and token requests. API keys do not work on the MCP server
3. Expect HTTP 429 with type rate_limit_error when the credit pool, the programmatic monthly cap or the key's spend cap is exhausted. Retrying will not help
4. Call GET /api/v1/w/{wId}/search with limit (1 to 100, default 25) and cursor, or a data source view's search with top_k
5. Keep document upserts under 120 a minute per workspace

Gemini Enterprise

1. Create a Gemini Enterprise app first. StreamAssist needs its APP_ID, and the caller needs discoveryengine.assistants.assist
2. Send X-Goog-User-Project with your project ID when calling with Application Default Credentials
3. Pass the session name from the final stream message on the next call to keep context. Leaving it empty starts a new session
4. Read connectorAuthErrors in the response. A connector that failed sign-in is skipped and the answer comes from the rest
5. Use the A2A URL returned by Agent Registry verbatim. It takes the project number, not the project ID