DeepInfra pricing
DeepInfra pricing — tracked: 1 listing for DeepInfra with current pricing, free credits and live status. Data-verified by UPROUTER.ONLINE.
$0 in free credits available
$ 1 entry matches — showing 1–1 · sort: Free credits ↓
388 routers & providers · 813 models tracked · $3.8k free credits
- Paid APIAPI keyunknownupdated 26d agodeepinfra.comsource · deepinfra.com/pricingoverview
DeepInfra is an OpenAI-compatible inference cloud for open-weight models, with per-token rates published on a single pricing page. Examples from that page: DeepSeek-V4-Flash-0731 at $0.06 per 1M input tokens ($0.015 cached) and $0.18 per 1M output; Qwen3.5-9B at $0.10/$0.15; DeepSeek-V3.2 at $0.26/$0.38; Kimi-K3 at $2.85 per 1M input and $14.25 per 1M output; embeddings from $0.005 per 1M input tokens. Models without fixed token rates are billed on execution time, and custom LLM deployments are billed per minute of GPU uptime across A100 ($0.89/GPU-hour) through B300 ($4.89/GPU-hour). There is no free credit: the pricing page states you must add a card or pre-pay before you can use the service.
free quotaNo free tier stated on official pages (checked 2026-09-16): deepinfra.com/pricing states that you must add a card or pre-pay before using the service, with no starter credit offered.
free modelsMeta: Llama 3.1 405B InstructMeta-Llama-3.1-70B-InstructLlama-3.1-8B-InstructQwen2.5 72B InstructDeepSeek ChatMixtral 8x22B Instructsee deepinfra.com’s free quota / risk review →
Every entry answers to the community
The index is expansive; the community keeps it honest. Votes, reviews and verified corrections feed straight back into rankings and data confidence.
- Vote
Upvote routers that deliver, downvote the ones that don’t — the score re-ranks the index.
- Review
Evidence-based reviews with verification tags. Ratings, not vibes.
- Verify
Corrections that survive review flip the entry to verified — with the evidence on file.
Anyone can edit this index
Every router page has an edit button — propose corrections to descriptions, prices and free tiers. Editors review each proposal against primary sources before it goes live.
- Propose
Open a router page, hit the edit button, and detail your proposed corrections.
- Review
Editors rigorously verify every proposed correction against primary sources.
- Live
Accepted edits are published instantly and logged in the public journal.
Recomputed from the live AI provider directory on every request — counts shift as providers change their tiers and pricing.
// the guide
DeepInfra pricing
5 min read · published October 2026 · updated October 2026
DeepInfra pricing — tracked: 1 listing for DeepInfra with current pricing, free credits and live status. Data-verified by UPROUTER.ONLINE.
Research and writing by the UPROUTER.ONLINE research team · verified against provider pricing pages on October 2026. Independently operated directory — no pay-to-rank, no affiliation with listed providers.
- The UPROUTER.ONLINE directory currently matches “DeepInfra pricing” to 1 listing — 1 with a free tier.
- Every row links to the provider's own pricing source and a verification date, and the counts below refresh live.
- Pricing rows are re-verified against the provider’s own pages; estimates and stale figures are flagged, never silently dropped.
DeepInfra pricing — tracked: 1 listing for DeepInfra with current pricing, free credits and live status. Data-verified by UPROUTER.ONLINE.
Written by Mike Fleming — Founder & Editor. He runs the directory, writes the entry research and reviews every page here; corrections go to the contact page.
Live sample — top listings
| Provider | Free credit value | Per-1M price | Risk | Status |
|---|---|---|---|---|
| DeepInfra | Not quantified | in $0 / out $0 | low | unknown |
DeepInfra pricing — the tracked numbers
On “deepinfra pricing”, the UPROUTER.ONLINE directory tracks 1 listing for DeepInfra — 1 with a free tier, 0 live in the latest status check.
Pricing rows are normalized to per-1M-token USD where a provider publishes rates; free allowances are converted at the provider's own pay-as-you-go rates and flagged as estimates. Figures that could not be verified are shown as “not quantified” rather than dropped or guessed.
Free credits vs pay-per-use
1 of the 1 tracked listing run a free tier. The tracked value is the allowance converted at the provider's own rates — a signup bonus, a monthly quota or a usage cap — so the column compares allowances, not lifetime spend.
Pay-per-use rows show the minimum input and output rates per 1M tokens in the directory. Where a rate could not be verified against the provider’s page, the field is left blank rather than estimated.
Risk, confidence and freshness
Each of the 1 entries carries a community risk rating (low / medium / high / unrated) with the evidence behind it, a data-confidence score, and a last-verified date per field. Status is probed continuously; the latest result is shown per entry.
Treat the rating as a starting point, not a clearance: the reasoning is public on each entry, and the provider’s own terms always win. Where data is stale or unverified, the directory marks it instead of hiding it.
Which one to pick
Start from what you actually need: a free allowance, a stable paid endpoint, or both. 1 of the 1 matching entries carry a free tier, so for most “deepinfra pricing” use cases there is a $0 starting point in the table above.
Narrow by risk tier and the last-verified date, then open the entry: the source URL, the terms and the status probe are the three things that matter. Ordering on this page is a data sort — free-credit value, then name — and nothing about it is sponsored. For DeepInfra specifically, the rows above are the complete match set, not a sample.
Verify any listing in four steps
1. Open the entry and read the source URL — the pricing, free-tier and access figures on the row come from the provider’s own page at that link.
2. Check the last-verified date and data-confidence score. Fresh, high-confidence rows are direct from the provider; older or community-reported rows are labeled as such. For DeepInfra, compare the row against the provider's current pricing page before you commit.
3. Read the risk rating and its reasoning. A low tier means established operator, clear terms and a stable history — a high tier means at least one of those is missing.
4. Confirm the live status probe result and the access method (api-key, subscription, local) match how you intend to use it. Then review the provider’s terms yourself — inclusion here is informational, never an endorsement.
How this page is built
Predicate — published, non-internal entries whose name, domain or description contains “deepinfra” (case-insensitive LIKE)
Sort — free-credit value descending, nulls last, provider name ascending (disclosed data sort — never sponsored)
Freshness — counts and rows are live from the dataset (60-second server cache); each row keeps its own per-field last-verified date, and unverified or stale values are flagged in place
Limitations — we track published pricing, free-tier terms and status — not account-level behavior. Free-credit figures are normalized estimates; risk is a community evidence rating, never a guarantee; inclusion is informational only.
Full methodology, probe cadence and the correction pathway: about & methodology · raw data: public dataset (CC BY 4.0) · pricing and status on every row are measurements with dates, not guarantees.
// sources & references
// faq
Frequently asked questions
How many “DeepInfra pricing” listings does the directory track?
1 published, non-internal entries currently match — 1 of them with a verified free tier. The figure is recomputed from the live dataset on every page load (60-second cache), so it moves as providers change their tiers.
Is “DeepInfra pricing” free?
Partly, and it's tracked: 1 of the 1 matching entries run a free tier. Each entry shows the shape — signup credit, monthly quota or free-only — plus the terms behind it, so “free” is never assumed.
How current is the “DeepInfra pricing” data on this page?
Counts, credit values and status are live from the dataset (cached for 60 seconds); each row also carries its own last-verified date per field. Where a figure went stale or unverifiable, the entry is marked rather than silently kept.
Who ranks the “DeepInfra pricing” listings?
No one, and nothing is sponsored. The ordering is a disclosed data sort — free-credit value descending, nulls last, name as tie-break — computed from the same rows the table renders.
Can I trust the risk ratings on “DeepInfra pricing” entries?
Treat them as community evidence ratings, not clearance. Each rating (low / medium / high) ships with its reasoning — operator age, terms transparency, reliability record — so you can audit the claim yourself against the source URLs.
Where does the pricing data come from?
Every row links to the provider's own pricing or terms page as the source of record. Free-tier values are normalized USD estimates converted at the provider's rates and flagged as estimates; unquantifiable tiers are shown as “not quantified”.
Do I need an account to use the “DeepInfra pricing” directory?
No. Browsing, filtering and reading every entry is free without login. An account only matters for community actions like voting and submitting corrections — and it never affects what data you can see.
What exactly does the “DeepInfra pricing” filter include?
The disclosed predicate in the methodology below: entries whose name, domain or description match, or that satisfy the deepinfra listings condition — no hand-picking, no exclusions beyond status and internal flags.