Fireworks AI through Squiid

What Fireworks AI is

Fireworks AI is an inference platform for open models. It serves chat, vision, embedding and image models over an OpenAI-compatible API, and adds production features on top: grammar and JSON schema constrained decoding so a response is guaranteed to parse, function calling, on-demand deployments for dedicated throughput, and fine-tuning that can be served without provisioning your own GPUs.

Constrained output is the reason Fireworks turns up in agent-written codebases. When a model's answer feeds straight into a database write or a UI component, a response that sometimes returns prose instead of JSON becomes a production bug. Teams route extraction, tagging and form-filling steps to Fireworks precisely because the schema is enforced at decode time rather than repaired afterwards. Fireworks also serves vision and image models, so a single provider can cover the OCR-style extraction step and the illustration step in the same application. The same argument applies to tool calling: a schema that is enforced rather than hoped for is the difference between an agent loop that finishes and one that silently stalls on a malformed argument.

Why connect it through Squiid

Squiid provisions Fireworks instantly and keeps the upstream key. Calls go to https://api.squiid.io/v1/fireworks-ai/... with Authorization: Bearer $SQUIID_API_KEY, and OpenAI-compatible clients work by pointing their base URL at the gateway (gateway convention). Usage is metered per model against prepaid credits at Fireworks' published per-token rates, dedicated deployments bill at their listed hourly rate on the same balance, and everything lands on one monthly Squiid invoice next to your database, hosting and email. One rotation in the Squiid dashboard invalidates the credential every agent on the team is using.

• One account, one key, one bill, one dashboard. Fireworks AI sits beside every other service your project uses, on the same invoice and the same credit balance.
• Key custody. Squiid holds the Fireworks AI credentials. Your code carries SQUIID_API_KEY, so FIREWORKS_API_KEY, FIREWORKS_BASE_URL 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 Fireworks AI 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

• Open-weight chat, vision and image models
• JSON schema constrained output through the gateway
• Fine-tuned model serving on the same key
• Usage and dedicated capacity on one balance
• Fireworks list pricing, no markup

How to connect Fireworks AI

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 FIREWORKS_API_KEY, FIREWORKS_BASE_URL
SQUIID_API_KEY=sq_live_xxxxxxxxxxxxxxxxxxxxxxxx

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

Pricing through Squiid

Fireworks charges per token by model size and per hour for dedicated deployments, at the provider's listed price. Squiid passes that through at par: one dollar of prepaid credit is one dollar of Fireworks AI usage, with no markup and no per-request margin.

Fireworks AI 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 Fireworks AI 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

Nothing to move: Fireworks AI is usage-metered, so leaving means creating your own Fireworks AI key and pointing the client at it. Squiid never owns your code, your schema or your domain, and the gateway speaks Fireworks AI'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

Fireworks AI 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 Fireworks AI 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.