● Per-token pricing
Pricing
You pay per token, in USD, with input and output priced separately. No minimums, no seats.
All prices
USD per 1M tokens. Prices last updated: 2026-09-26.
| Model | Creator | Type | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| BGE-M3 | BAAI | embedding | 8K | $0.01 | — |
| Qwen3 Embedding 8B | Qwen | embedding | 32K | $0.02 | — |
| Mistral Small 3.2 | Mistral | chat | 128K | $0.06 | $0.18 |
| Gemma 3 27B | chat | 128K | $0.09 | $0.17 | |
| Qwen3 32B | Qwen | chat | 40K | $0.10 | $0.30 |
| Llama 3.3 70B | Meta | chat | 128K | $0.23 | $0.40 |
| DeepSeek V3 | DeepSeek | chat | 128K | $0.27 | $1.10 |
| DeepSeek R1 | DeepSeek | chat | 128K | $0.55 | $2.19 |
FAQ
How am I billed?
Each request is charged for the input tokens you send and the output tokens the model returns, at the per-1M-token prices above. Payment options are shown in the console.
Is there a minimum spend?
No. You pay only for the tokens you use.
Which SDKs work?
Any OpenAI-compatible SDK or tool. Set the base URL to https://api.keyra.example/v1 and use your Keyra API key.
Do prices change?
Prices are published on this page. The date above shows when they last changed.