Pay per token, at the price shown. Add funds once and use any model with a single API key — no plan to pick, nothing charged until you make a request.
Top up by card or crypto. Every model draws from the same balance.
A single API key works across the whole catalogue. Switch models without changing anything.
Your thread carries across models. Start with one, continue with another.
| Model | Input / 1M tokens | Output / 1M tokens | Context |
|---|---|---|---|
| auto alphaneural | varies — you pay the price of whichever model answers | 1M | |
| nova-2-lite amazon | $0.363/1M | $3.02/1M | 1M |
| nova-lite amazon | $0.066/1M | $0.264/1M | 300K |
| nova-micro amazon | $0.038/1M | $0.154/1M | 128K |
| nova-pro amazon | $0.880/1M | $3.52/1M | 300K |
| claude-haiku-5.5 anthropic | $0.121/1M | $0.605/1M | 200K |
| claude-opus-5 anthropic | $6.05/1M | $30.25/1M | 200K |
| claude-opus-5.5 anthropic | $4.84/1M | $24.20/1M | 200K |
| claude-sonnet-5 anthropic | $2.42/1M | $12.10/1M | 200K |
| claude-sonnet-5.5 anthropic | $2.42/1M | $12.10/1M | 200K |
| deepseek-r1 deepseek | $1.48/1M | $5.94/1M | 64K |
| deepseek-v3.2 deepseek | $0.682/1M | $2.04/1M | 164K |
| gemma-3-27b-it google | $0.253/1M | $0.418/1M | 131K |
| llama-3.3-70b-instruct meta-llama | $0.792/1M | $0.792/1M | 128K |
| minimax-m2.5 minimax | $0.330/1M | $1.32/1M | 205K |
| devstral-2 mistralai | $0.440/1M | $2.20/1M | 262K |
| magistral-small-2509 mistralai | $0.550/1M | $1.65/1M | 128K |
| mistral-large-3 mistralai | $0.550/1M | $1.65/1M | 262K |
| pixtral-large-2502 mistralai | $2.20/1M | $6.60/1M | 128K |
| kimi-k2-thinking moonshotai | $0.660/1M | $2.75/1M | 262K |
| kimi-k2.5 moonshotai | $0.660/1M | $3.30/1M | 262K |
| kimi-k3 moonshotai | $3.63/1M | $18.15/1M | 262K |
| nemotron-super-3-120b nvidia | $0.165/1M | $0.715/1M | 262K |
| gpt-5.6-luna openai | $0.242/1M | $1.45/1M | 400K |
| gpt-5.6-sol openai | $4.84/1M | $24.20/1M | 400K |
| gpt-6-luna openai | $0.121/1M | $0.605/1M | 400K |
| gpt-6-sol openai | $2.42/1M | $12.10/1M | 400K |
| gpt-6.1-sol openai | $2.42/1M | $12.10/1M | 400K |
| gpt-oss-120b openai | $0.165/1M | $0.660/1M | 131K |
| gpt-oss-20b openai | $0.077/1M | $0.330/1M | 131K |
| qwen3-coder-next qwen | $0.550/1M | $1.32/1M | 262K |
| qwen3-next-80b-a3b qwen | $0.154/1M | $1.32/1M | 262K |
| qwen3-vl-235b-a22b qwen | $0.583/1M | $2.93/1M | 262K |
| grok-4.6 x-ai | $2.42/1M | $7.26/1M | 500K |
| grok-4.7 x-ai | $2.42/1M | $7.26/1M | 500K |
| glm-4.7 z-ai | $0.660/1M | $2.42/1M | 205K |
| glm-5 z-ai | $1.10/1M | $3.52/1M | 205K |
Prices are per million tokens and include everything. What you are billed is the exact cost of each request as the provider reports it; a few models are priced by time of day, and image, audio and other non-text usage is billed at the provider’s rate for it.