20 models across 4 providers, all behind one key. Rates are read from the gateway as you load this page.
| Model | Provider | Modality | Context | Input | Output | Cached input |
|---|---|---|---|---|---|---|
| claude-fable-5-1claude-fable-5-1 | anthropic | text | 1.0M | $10 / M | $50 / M | $0.250 / M |
| claude-haiku-4-5claude-haiku-4-5 | anthropic | text | 200k | $1.00 / M | $5.00 / M | $0.100 / M |
| claude-haiku-4-5-20251001claude-haiku-4-5-20251001 | anthropic | text | 200k | $1.00 / M | $5.00 / M | $0.100 / M |
| claude-opus-5claude-opus-5 | anthropic | text | 1.0M | $5.00 / M | $25 / M | $0.500 / M |
| claude-sonnet-5claude-sonnet-5 | anthropic | text | 1.0M | $2.00 / M | $10 / M | $0.200 / M |
| gpt-oss-120bcerebras/gpt-oss-120b | cerebras | text | 131k | $0.350 / M | $0.750 / M | — |
| Google Gemini 2.5 Flashgemini-2.5-flash | gemini | text | 1.0M | $0.075 / M | $0.300 / M | — |
| GPT-4.1 minigpt-4.1-mini | openai | text | 1.0M | $0.400 / M | $1.60 / M | $0.100 / M |
| GPT-5.6 Lunagpt-5.6-luna | openai | text | 400k | $0.200 / M | $1.20 / M | $0.020 / M |
| gpt-4o-mini-transcribegpt-4o-mini-transcribe | openai | audio | 16k | $1.25 / M | $5.00 / M | — |
| gpt-4o-transcribegpt-4o-transcribe | openai | audio | 16k | $2.50 / M | $10 / M | — |
| gpt-6-astragpt-6-astra | openai | text | 922k | $10 / M | $50 / M | $1.00 / M |
| gpt-image-2gpt-image-2 | openai | image | — | $5.00 / M | $30 / M | $1.25 / M |
| gpt-image-2.5-flaregpt-image-2.5-flare | openai | image | — | $5.00 / M | $30 / M | $1.25 / M |
| gpt-image-2.5-sunburstgpt-image-2.5-sunburst | openai | image | — | $5.00 / M | $30 / M | $1.25 / M |
| gpt-transcribegpt-transcribe | openai | audio | — | — | — | — |
| Reprompt Scribe v1reprompt-scribe-v1 | — | audio | — | — | — | — |
| text-embedding-3-smalltext-embedding-3-small | openai | embedding | 8k | $0.020 / M | $0.000 / M | — |
| whisper-1whisper-1 | openai | audio | — | — | — | — |
| Qwen3.8 27B (Cerebras)cerebras/qwen-3.8-27b | cerebras | text | 131k | $0.400 / M | $0.800 / M | — |
Rates are cents per million tokens as the gateway prices them today. Cached rates apply where the provider supports prompt caching.