DeepInfra
Low-cost serverless inference for open models with transparent per-token rates.
DeepInfra FLUX Pro image
$0.05 / image
DeepInfra Voxtral Mini speech-to-text
$0.001 / audio minute
About DeepInfra
Low-cost serverless inference for open models with transparent per-token rates. DeepInfra sits in CostUse’s catalog of usage-billed APIs. Instead of a thin brand blurb, this page assembles what matters for budgeting: what the company bills for usage, in which unit, at which list rate, and on which date the figure was checked against the public source.
There are 2 published calculators for DeepInfra. List unit prices run from $0.001 to $0.05, with a median of $0.0255. Billing units represented here: image, audio minute. The most recent price check in this group is 2026-07-30.
Official DeepInfra pricing page
DeepInfra pricing table
Compare meters side by side before opening each calculator. Watch the unit carefully: a “cheap” host rate is not comparable to a per-GB or per-million-events rate. The source column links to the official page used for verification.
| Service | Unit price | Unit | Source | Verified |
|---|---|---|---|---|
| DeepInfra Voxtral Mini speech-to-text | $0.001 | audio minute | DeepInfra pricing | |
| DeepInfra FLUX Pro image | $0.05 | image | DeepInfra pricing |
DeepInfra services and APIs in this catalog
Each table row maps to a full service page: volume calculator, cost profile (what 1, 1k, and 1M units cost, and how far $100 goes), and the pricing documentation link. If DeepInfra publishes included quotas on monthly plans, use the calculator for the variable part and read the inclusion on the official source.
For engineering and finops teams, the practical flow is: (1) list the meters your product actually hits, (2) fill real or peak volume, (3) sum the results instead of guessing a single “platform average.” Large platforms mix several counters—logs, APM, infrastructure, RUM—and budget misses start when every line is treated as the same meter.
DeepInfra calculators
The cards at the top of this page open each service with its interactive calculator. Start with the meter that dominates your current bill, then explore the rest of the DeepInfra lineup.
Frequently asked questions about DeepInfra
What does DeepInfra bill for usage in this catalog?
There are 2 published meters for DeepInfra, billed in units such as image, audio minute. Each page shows the list unit price, the billing unit, and the date the figure was checked against the public source.
Which DeepInfra meter has the lowest unit price here?
DeepInfra Voxtral Mini speech-to-text lists $0.001 per audio minute. Use that page’s calculator to project daily and monthly volume before you lock a budget.
Do DeepInfra monthly plans roll into the calculator figure?
When the source publishes a per-unit rate (per GB, per request, per seat, and so on), the calculator uses that linear list rate. Included quotas, monthly minimums, annual packs, and volume discounts still need to be read on the official docs—CostUse keeps the source and verification date visible on every service page.
How should I compare DeepInfra services against each other?
Use the pricing table on this page and sort by what you actually consume: log ingest, hosts, spans, seats, and storage are not comparable on the same unit. Then run each calculator with your team’s real volume, not a generic placeholder.