Usage-based pricing

Top up credits. Successful training runs debit credits based on training tokens, epochs, and model rate.

Add prepaid credits before your first run. Failed runs are free.

Rate card

Price per 1M training tokens, per epoch.

ModelSizePrice / 1M tokens · epoch
Qwen/Qwen3.5-2B2B$0.45
google/gemma-4-E2B-itE2B$0.45
ibm-granite/granite-3.3-2b-instruct2B$0.45
meta-llama/Llama-3.2-3B-Instruct3B$0.55
bigcode/starcoder2-3b3B$0.55
Qwen/Qwen3.5-4B4B$0.70
google/gemma-4-E4B-itE4B$0.70
microsoft/Phi-4-mini-instruct3.8B$0.70

Rates may change as provider pricing changes. Check the docs for billing details.

Top up before your first run

No subscription. Pay from day one with prepaid credits.