Turbopuffer through Squiid

What Turbopuffer is

Turbopuffer takes a different architectural bet from other vector databases: indexes live on object storage rather than in memory, with caching in front for hot data. The result is that storing a very large number of vectors costs roughly what storing files costs, and you pay for queries when they happen instead of paying to keep everything resident in RAM. It supports vector search, full-text search and filtering over namespaces, with writes that are durable as soon as they are acknowledged. Namespaces are created implicitly on first write, so a per-customer index does not need a provisioning step of its own.

This shape fits a specific and common situation in AI-coded products: a corpus that is huge but cold. Per-customer document archives, historical email or ticket data, logs that are occasionally searched, and long-tail knowledge bases are all expensive on memory-resident databases and cheap here. The trade is latency on a cold namespace, which matters much less when the query is part of a chat response that already takes a second. It is also a reasonable default for an app that indexes everything by habit, because storing data you rarely search is cheap enough not to think about.

Why connect it through Squiid

Turbopuffer is on the Squiid roadmap. When it is connected, Squiid will provision the account, hold the API key and route requests through https://api.squiid.io/v1/turbopuffer/... with Authorization: Bearer $SQUIID_API_KEY, with clients working by base URL swap (gateway convention). Storage, writes and queries will meter against prepaid credits at Turbopuffer's listed price on one monthly invoice, so the cost of an archive that grows every day is visible next to the model and storage spend that feeds it.

• One account, one key, one bill, one dashboard. Turbopuffer sits beside every other service your project uses, on the same invoice and the same credit balance.
• Key custody. Squiid holds the Turbopuffer credentials. Your code carries SQUIID_API_KEY, so TURBOPUFFER_API_KEY, TURBOPUFFER_REGION never land in a repository, a preview deployment or an agent’s context.
• Metering you can see. Every call is counted per service and per day, so you know what Turbopuffer costs before the month ends.
• Pause on zero. When prepaid credits run out the gateway stops authorizing usage instead of running up a bill you did not fund.
• One rotation. Rotating the Squiid key cuts off every agent, worker and deployment at once, with no provider key to hunt down.

What you get

• A provisioned Turbopuffer account
• Vector and full-text search over namespaces
• Object-storage economics for large cold corpora
• Queries and storage metered against credits
• Turbopuffer list pricing, no markup

How to connect Turbopuffer

One variable replaces the provider credentials this service would normally need. Requests go to the gateway and the official SDKs work by pointing their base URL at it (gateway convention).

# .env
# replaces TURBOPUFFER_API_KEY, TURBOPUFFER_REGION
SQUIID_API_KEY=sq_live_xxxxxxxxxxxxxxxxxxxxxxxx

# every request: https://api.squiid.io/v1/turbopuffer/...
# header:       Authorization: Bearer $SQUIID_API_KEY

Pricing through Squiid

Turbopuffer charges for stored data, writes and queries, at the provider's listed price. Squiid passes that through at par: one dollar of prepaid credit is one dollar of Turbopuffer usage, with no markup and no per-request margin.

Turbopuffer has no subscription attached on Squiid today. If a plan is required for a feature you need, it passes through at the provider’s listed price with no Squiid fee.

Squiid adds no markup to provider prices. There is a free plan at $0 a month with a 15% fee when you top up credits, Solo at $19.97 a month with a top-up fee that starts at 8% and falls to 3.5% as your spend grows, Team from $29.97 a seat a month with the same ladder one point lower, and Custom with a rate of 3% or less. Credits are prepaid, so Turbopuffer usage stops when the balance reaches zero rather than producing an invoice you did not fund. Auto top-up with a cap keeps production running without handing anyone an open-ended card.

If you leave

Export: records, vectors and index settings are exported and re-imported into an index in your own Turbopuffer account. Squiid never owns your code, your schema or your domain, and the gateway speaks Turbopuffer's own API, so the client change is a base URL. The handover page lists what happens for every provider in the catalogue.

Works with every coding agent

Turbopuffer through Squiid is just an HTTP call with a bearer token, so every agentic coding tool can use it with the single key already in your environment:

• Claude Code
• Cursor
• ChatGPT
• Codex
• Grok
• Gemini CLI
• GitHub Copilot
• Windsurf
• Perplexity
• Devin
• Replit
• v0
• Lovable
• Cline

Vibe coding a feature on Turbopuffer usually means asking for it in plain language and letting the agent write the integration. Because the key is Squiid’s rather than the provider’s, it can do that without ever holding a credential that could be leaked, logged or committed.