AskCat Daily: PRICE HIKE, TIER CUT — we'll warn you first
This tool's price check is 25 days old: page was read but no Taiwan price could be matched, not stamped. The prices below may be out of date, so please confirm on the official page. We do not stamp a fresh date on it. Check the official page
Auto-admitted challenger candidate: data checked once, not yet triple-verified by a human. Treat the free allowance and commercial terms as whatever the official site says today.
View Change History
2026-10-25~2026-11-19Routine checks ran 2 times; no price or plan changes
2026-10-26~2026-11-19Automated checks could not be matched line by line with the official site 24 times (page redesign, region-based pricing or usage-based billing); not judged to be a price change; the verified date was not updated, pending manual review
Continuously re-checked, every fact dated
Has a free plan
What it is
Fast, production-ready inference API and fine-tuning platform for state-of-the-art open-source generative AI models.
How far the free plan goes
Newly registered users receive a US$1 free trial credit, and the free account has a limit of 10 requests per minute (10 RPM).
Free plan limit
Features
Free quota
Description
Free-tier allowance
Newly registered users receive a US$1 free trial credit, and the free account has a limit of 10 requests per minute (10 RPM).
As seen on the official site on the day of entry; see the official site for details
Charges are based on token usage: <4B‑parameter models are approximately US$0.10/M tokens, 4B‑16B models about US$0.20/M tokens, and >16B models start at roughly US$0.90/M tokens; cached inputs receive a 50% discount, and batch inference (Batch) is offered at half price.
Which Model
Fireworks AI
Usage Quota
Charges are based on token usage: <4B‑parameter models are approximately US$0.10/M tokens, 4B‑16B models about US$0.20/M tokens, and >16B models start at roughly US$0.90/M tokens; cached inputs receive a 50% discount, and batch inference (Batch) is offered at half price.
It primarily uses prepaid pay-as-you-go billing (Serverless API) and dedicated GPU rentals (hourly billing, H100/H200 starting at about US$7‑8 per hour). It supports LoRA/Full SFT model fine‑tuning (starting at about US$0.50‑US$10 per million training tokens). After binding a credit card, inference speed can increase to 6,000 RPM.
Looking for Fireworks AI deals, discounts or promo codes? We check the official site automatically every day, review changes by hand, and date-stamp every entry.