Transparent · Pay per token
LLM API pricing across 785 models
LLM API pricing on FastInfra is per token: every model lists a live input price and output price in USD per million tokens, synced from upstream providers and shown before you spend anything. Free-tier chat models cost $0 within their rate limits. Paid models draw from prepaid credits with no subscription, no monthly minimum and no per-seat fee, and FastInfra routes each request to the lowest-cost healthy route for that model.
Search, sort, and deploy in minutes. No hidden fees — one unified price per model. Catalog updated Oct 10, 2026.
One $5 top-up unlocks every prepaid model. No subscription, credit never expires. Before you pay, check live latency and uptime, the compatibility matrix, and published rate limits that only go up.
No models match your search.
| Gemma2:2b gemma2:2b | $0 | $0 |
| Llama3.1:8b llama3.1:8b | $0 | $0 |
| Llama3.2:1b llama3.2:1b | $0 | $0 |
| Llama3.2:3b llama3.2:3b | $0 | $0 |
| Mistral:7b mistral:7b | $0 | $0 |
| Mxbai Embed Large mxbai-embed-large | $0 | $0 |
| Nomic Embed Text nomic-embed-text | $0 | $0 |
| Phi3:mini phi3:mini | $0 | $0 |
| Qwen2.5:0.5b qwen2.5:0.5b | $0 | $0 |
| Qwen2.5:1.5b qwen2.5:1.5b | $0 | $0 |
| Qwen2.5:14b qwen2.5:14b | $0 | $0 |
| Qwen2.5:3b qwen2.5:3b | $0 | $0 |
Frequently asked questions
How is LLM API usage priced on FastInfra?
Per token. Each model shows an input price and an output price in USD per 1 million tokens. Your request is billed as prompt tokens multiplied by the input price plus completion tokens multiplied by the output price.
Why do input and output tokens have different prices?
Generating a token costs more compute than reading one, so providers charge more for output. Output prices are typically two to five times the input price on the same model, which is why long responses dominate most bills.
Is there a subscription or minimum spend?
No. You top up prepaid credits and usage draws them down. There is no monthly plan, no seat fee and no minimum commitment; free-tier models stay at $0 within their limits.
How often do prices change?
Prices sync from upstream providers automatically. The Pricing page and every model page always show the current rate, and a cheaper healthy route for the same model is used automatically when one exists.
Do free-tier models have limits?
Yes. Free-tier models have published requests-per-minute and concurrency limits that rise after a first top-up. The limits are listed in the API documentation under Rate Limits.
Last reviewed October 15, 2026 by FastInfra. Prices and counts on this page come from the live catalog.
Start shipping today
785 models. One API key. Zero integration overhead. Create your account and make your first request in under five minutes.