Tools Web

Calculators

Novita Token Cost Calculator

Estimate Novita Model API cost from ~12 curated presets (DeepSeek V4.1 Flash / V4 Pro, GLM 5.3 Flash/5.3, Kimi K3.1 Pro cache-read simplified, Qwen3.8 Flash, MiniMax M3.1 Flash, MiMo V2.6 Flash, Llama 4 Scout, Gemma 4 31B). Optional Batch introductory 50% discount on input/cached/output. Not the 200+ catalog; GPU Instance and Agent Sandbox omitted. Kimi cache-write TTLs noted in FAQ only. Estimates only; official Novita pricing prevails.

Math runs in your browser. Nothing is uploaded.

Input cost
—

Cached input cost
—

Output cost
—

Estimated USD cost
—

Monthly USD
—

Estimates only; official vendor pricing prevails (2026-10-30). Cost = uncached in÷1M×rate + cached÷1M×rate + out÷1M×rate; optional Batch ×0.5. Curated ~12 presets (not 200+ catalog). Kimi K3.1 Pro cache-write 5m/1h noted in FAQ only. No GPU Instance / Agent Sandbox.Official pricing: Novita pricing

DeepInfra Token Cost Calculator · Hyperbolic Token Cost Calculator · SiliconFlow Token Cost Calculator · Fireworks Token Cost Calculator · OpenRouter Token Cost Calculator · Together AI Token Cost Calculator

More in this group

How it works

What does Batch 50% do?
Introductory Batch discount applies 50% to input, cached, and output rates for supported models. Confirm current eligibility on novita.ai/pricing.
Why is Kimi cache-write simplified?
Kimi K3.1 Pro publishes cache-write 5m/1h rates ($2/$4). This page models cache-read as the cached field; add write tokens manually via rate fields or use the Kimi calculator for full TTL accounting.
Full 200+ catalog?
No. v1 is curated presets only — not GPU Instance or Agent Sandbox. Paste custom $/Mt into the rate fields if needed.
Are results invoices?
No. Estimates only; novita.ai/pricing prevails.
Is any data uploaded?
No. These calculators and QR codes run in your browser.