> Novita AI
Novita AI runs 200+ models across text, image, audio and video behind an OpenAI-compatible API at api.novita.ai/openai, plus serverless endpoints, GPU instances and agent sandboxes, all billed per token or per GPU-second. Published model rates start at $0.02/1M input and $0.05/1M output for Llama 3.1 8B Instruct, with DeepSeek V4 Flash at $0.14/$0.28 on a 1M-token context, and three Ling 3.0 Flash variants (Fin, VL, Sante) listed at Free input and Free output with 256K context. Batch inference carries an introductory 50% discount on input and output tokens for supported models.
// pros
- Three models carry $0 input and $0 output pricing on the public table: Ling 3.0 Flash Fin, Ling 3.0 Flash VL and Ling 3.0 Flash Sante, all at 256K context.
- Lowest metered rates published are Llama 3.1 8B Instruct at $0.02/1M input and $0.05/1M output, with DeepSeek V4 Flash at $0.14/$0.28 and cache reads at $0.028/1M.
- Batch inference is at an introductory 50% discount on input and output tokens for supported models, per the note on the pricing page linking to the batch API guide.
- Prompt-cache reads are priced in the same row as standard input (for example $0.006/1M on DeepSeek V4.1 Flash against $0.30/1M input), so cached workloads can be costed before deployment.
- One key reaches text, image, audio and video models, and the docs list official integrations for LangChain, LiteLLM, Dify, Cursor and the OpenAI Agents SDK.
- OpenAI-compatible base URL (api.novita.ai/openai) means existing OpenAI client code works with a key and base-URL change.
// cons
- No standing USD sign-up credit appears on the current pricing or docs pages; the $10 new-user offer was a limited-time blog promotion, and a September 2026 official post answers 'Is it free? No.' for hosted models.
- The batch discount is labelled introductory, so effective rates for batch workloads can revert while the headline price on the table stays the same.
- The three $0 rows are the only free entries; every DeepSeek and Llama model is metered, so the free path is limited to one narrow model family.
- The catalogue carries many dated near-duplicates with very different prices (DeepSeek V4 Flash at $0.14/$0.28 versus DeepSeek V4 Flash 0731 at $0.44/$1.32), so cost depends on pinning the exact model id.
- The pricing page does not state expiry, quota or rate limits for the $0-priced models, so free capacity cannot be planned against.
Use Novita when model breadth and low per-token rates matter, because image, video and audio models sit behind the same key as text, and start on a free Ling 3.0 Flash variant before committing spend. In a router it works well as a low-cost secondary provider for DeepSeek V4 Flash and Llama-class traffic.
written by Mike Fleming, Uprouter editorial · verified against official sources September 16, 2026
Free tier present — value not quantified
This provider offers a free tier, but its terms don’t convert cleanly into a USD figure (e.g. request-count limits or capacity-dependent pools). We refuse to print a misleading zero here. See the note below and the official pricing page for the exact limits.
note: No standing free credit is stated on the current official pricing or docs pages (checked 2026-09-16); the pricing table lists Ling 3.0 Flash Fin/VL/Sante at Free input and output (256K context), and an official blog post from May 2025 advertised a limited-time $10 new-user credit that current pages do not repeat.
| Plan | Type | Monthly | Input /1M | Output /1M | Note |
|---|---|---|---|---|---|
| Free tier (source-reported) | free | — | — | — | New users get about $0.50 in free credit for text/image API testing (~60 RPM); a referral program gives $10 each to referrer and referee, up to $500 total. Pay-as-you-go once used up. |
| Free tier | free | — | Free | Free | ~$0.50 one-time |
| Pay-as-you-go | payg | — | $0.400 | $0.400 | deepseek-r1-distill class |
| DeepSeek V4 Flash | payg | — | $0.140 | $0.280 | Ultra-competitive rates for the latest reasoning models. |
| Llama 3.3 70B | payg | — | $0.135 | $0.400 | — |
Pricing normalized from public sources — always verify with the provider.
// model prices
| Model | Input /1M | Output /1M | Context |
|---|---|---|---|
| DeepSeek family | — | — | — |
| Qwen family | — | — | — |
| Llama family | — | — | — |
$ Does Novita AI have a free tier?
Yes — Novita AI offers a free tier. Always confirm current limits on Novita AI's official pricing page — free tiers change without notice.
$ How much does Novita AI cost per 1M tokens?
Novita AI does not publish simple per-1M-token pricing (it may bill per request, per output, or via a subscription plan). Check their official pricing page for exact figures.
$ Is Novita AI safe to use?
Uprouter rates Novita AI as low risk. Medium. Nothing on the official pricing or docs pages grants a standing free credit, and the $0-priced rows carry no published limits, so free access should be treated as best-effort rather than capacity. Promotional pricing is a pattern here: the batch 50% discount is labelled introductory and the $10 new-user credit was announced for a limited time, so budget figures need re-checking against the live pricing page. The terms-of-service URL checked during this review returned not found, so account-level data-use terms could not be verified here. Risk ratings are editorial, evidence-based, and never influenced by affiliate relationships.
Raw community sentiment, shown unblended. Votes never feed into objective data fields like price, risk, or status — and never into sort order.
No published reviews yet — be the first.
// more_informationprice history · live status · risk & confidence · code snippets
price_history
No recorded changes yet.
live_status
full history →best-effort probes · not a guarantee · last 15h ago
| Checked | Status | Latency |
|---|---|---|
| 15h ago | up | 200ms |
| 1d ago | up | 225ms |
| 3d ago | up | 216ms |
| 4d ago | up | 278ms |
| 6d ago | up | 315ms |
| 7d ago | up | 233ms |
| 9d ago | up | 265ms |
| 10d ago | up | 200ms |
| 12d ago | up | 243ms |
| 13d ago | up | 700ms |
| 14d ago | up | 239ms |
| 15d ago | up | 248ms |
| 15d ago | up | 256ms |
| 16d ago | up | 268ms |
| 17d ago | up | 325ms |
| 18d ago | up | 220ms |
| 18d ago | up | 320ms |
| 19d ago | up | 390ms |
| 20d ago | up | 391ms |
| 21d ago | up | 651ms |
| 21d ago | up | 297ms |
| 22d ago | up | 250ms |
| 23d ago | up | 284ms |
| 24d ago | up | 297ms |
| 25d ago | up | 269ms |
| 26d ago | up | 269ms |
| 26d ago | up | 314ms |
| 27d ago | up | 534ms |
| 28d ago | up | 450ms |
| 29d ago | up | 495ms |
Uprouter runs its own probes; best-effort, not a guarantee. Probe results depend on our network location and cadence — for production, always use the provider’s own status page.
risk_&_confidence
Medium. Nothing on the official pricing or docs pages grants a standing free credit, and the $0-priced rows carry no published limits, so free access should be treated as best-effort rather than capacity. Promotional pricing is a pattern here: the batch 50% discount is labelled introductory and the $10 new-user credit was announced for a limited time, so budget figures need re-checking against the live pricing page. The terms-of-service URL checked during this review returned not found, so account-level data-use terms could not be verified here.
Confidence reflects how much verifiable evidence (official docs, probe history, community reports) backs this entry. Risk is our editorial assessment of operator reliability and terms — not financial advice.
code_snippets
# Novita AI's own API — get a key from their dashboard
curl https://YOUR-PROVIDER-ENDPOINT/chat/completions \
-H "Authorization: Bearer THEIR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-family",
"messages": [{ "role": "user", "content": "Hello" }]
}'Direct to the provider — Uprouter receives nothing; verify their current pricing before purchasing.
Ranked by neutral similarity signals only — live status, free-tier shape and free-credit proximity. Commercial relationships never influence this list.