The cheapest AI tokens, fixed in USD

Same open-weight models, a fraction of the price. Transparent per-million-token rates, no opaque credits, no subscriptions.

Priced to be the cheapest

Open-weight models on cost-efficient infrastructure at a fixed USD margin — roughly 4× less per token than OpenRouter on the same models. You see the exact rate before you spend.

Cheap and drop-in

Cheap doesn't mean different — it's the same OpenAI/Anthropic API and the same models, just a different base URL and key. Pay-as-you-go from a USD balance topped up with crypto.

Drop-in: change the base URL

env
# The cheapest drop-in for OpenAI/Anthropic clients
export OPENAI_BASE_URL="https://api.cheaptokens.dev/v1"
export OPENAI_API_KEY="YOUR_CHEAPTOKENS_KEY"

Priced below the rest

Fixed per-million-token USD rates versus OpenRouter on the same open-weight models.

Model OpenRouter cheaptokens You save
GLM-5.3-Flash $0.048 / $0.20 $0.027 / $0.08 ~2× cheaper
DeepSeek-V4-Flash $0.016 / $0.64 $0.0221 / $0.10 ~4× cheaper

Same models. USD per 1M tokens (input / output). OpenRouter shown for comparison; savings by blended price (3 parts input to 1 part output). · Open cost calculator

FAQ

Are these really the cheapest tokens?

Fixed USD per-token on open-weight models, far below first-party APIs — see the comparison above.

Any hidden fees?

No — published per-model rates and a small deposit fee; no subscription.

How do I start?

Create an account, top up with crypto, get a key, and point your client at our base URL.

Get the cheapest tokens

Create an account, top up with crypto, and start spending from a USD balance.

Get started