← All providers
Novita AI pricing
verified 2026-09-02Broad low-cost catalog of open-weight models on serverless endpoints.
data residency & retention
from Novita AI's own policies · verified 2026-09-02 · some facts not publishedHeadquarters
Not disclosed
Where inference runs
Not disclosed
Novita says only that it partners with datacenters around the world; no specific datacenter countries are disclosed, and the privacy policy permits cross-border transfers including to the United States.
Region pinning
Not disclosed
No region-pinning documented. Novita also does not disclose its controller entity or HQ — the terms are governed by Delaware (US) law, while third-party listings variously cite San Francisco, Hong Kong, or Singapore (unverified).
Prompt retention
Not retained
The terms state content is not retained beyond the time it takes to generate and deliver output, except as required by law; account and transaction data is kept for multi-year periods.
Trains on API data
No
By default Novita will not use customer content to train its own models or improve the services.
Zero data retention
Available
Zero data retention is described in the terms as the default policy for content rather than an opt-in feature.
current models · $ per 1M tokens
| Model | Input /1M | Cached in | Output /1M | Context |
|---|---|---|---|---|
OpenAI GPT OSS 120B GPT-OSS | $0.05 | — | $0.25 | 131K |
GLM 5.3 Flash GLM | $0.07 | — | $0.25 | 1.0M |
DeepSeek V4 Flash 0731 DeepSeek | $0.44 | — | $1.32 | 1.0M |
Llama 4 Maverick Instruct Llama | $0.27 | — | $0.85 | 1.0M |
DeepSeek V4 Flash DeepSeek | $0.14 | — | $0.28 | 1.0M |
DeepSeek V4 Pro DeepSeek | $1.60 | — | $3.20 | 1.0M |
DeepSeek V4 Pro 0813 DeepSeek | $1.32 | — | $3.96 | 1.0M |
Qwen3.8 27B Qwen | $0.42 | — | $3 | — |
DeepSeek V3.2 DeepSeek | $0.27 | — | $0.40 | 164K |
Kimi K3 Kimi | $3 | — | $15 | 1.0M |
MiniMax M2.7 MiniMax | $0.30 | — | $1.20 | 205K |
Novita AI totals · 1M in + 1M out
OpenAI GPT OSS 120B
$0.3000
GLM 5.3 Flash
$0.3250
DeepSeek V4 Flash
$0.4200
DeepSeek V3.2
$0.6690
Llama 4 Maverick Instruct
$1.12
MiniMax M2.7
$1.50
DeepSeek V4 Flash 0731
$1.76
Qwen3.8 27B
$3.42
DeepSeek V4 Pro
$4.80
DeepSeek V4 Pro 0813
$5.28
Kimi K3
$18.00
Cost of 1M input + 1M output tokens. Bar length is square-root scaled so cheap models stay visible next to premium ones.