Novita AI models
Open-model serverless API with competitive per-token rates across Llama, Qwen, and DeepSeek.
Llama 3.1 8B Instruct (Novita)
$0.02 / $0.05 per 1M tokens
Llama 3.3 70B Instruct (Novita)
$0.135 / $0.4 per 1M tokens
Llama 4 Scout (Novita)
$0.18 / $0.59 per 1M tokens
Llama 4 Maverick (Novita)
$0.27 / $0.85 per 1M tokens
DeepSeek V3.2 (Novita)
$0.269 / $0.4 per 1M tokens
Novita AI
Open-model serverless API with competitive per-token rates across Llama, Qwen, and DeepSeek.
There are 5 published models here. On a 70/30 mix, combined list rates run from $0.029 to $0.444 per 1 million tokens, with a median of $0.303. The largest listed context window in this group is 1,048,576 tokens. The most recent price check in this group is 2026-08-14.
Official Novita AI pricing page
Novita AI list rates
USD per million tokens. The 70/30 mix is a planning default — change it on the model page if your traffic is output-heavy.
| Model | Input / 1M | Output / 1M | Mixed 70/30 | Context | Verified |
|---|---|---|---|---|---|
| Llama 3.1 8B Instruct (Novita) | $0.02 | $0.05 | $0.029 | 16,384 | |
| Llama 3.3 70B Instruct (Novita) | $0.135 | $0.4 | $0.2145 | 128,000 | |
| Llama 4 Scout (Novita) | $0.18 | $0.59 | $0.303 | 131,072 | |
| DeepSeek V3.2 (Novita) | $0.269 | $0.4 | $0.3083 | 160,000 | |
| Llama 4 Maverick (Novita) | $0.27 | $0.85 | $0.444 | 1,048,576 |
Questions about Novita AI rates
Do these Novita AI figures include discounts and tax?
No. They are public list rates. Commits, credits, regions, tax, and commercial discounts are not in the number — check the source linked on each model.
When is the more expensive Novita AI tier worth the extra?
Llama 4 Maverick (Novita) sits near $0.444 per 1M tokens on a 70/30 mix, versus $0.029 on Llama 3.1 8B Instruct (Novita). Use the high tier when retries or long reasoning actually fail on the cheap one.