OpenLuxEvery model provider, one API key
OpenLux serves 9 model vendors behind a single OpenAI-compatible endpoint: OpenAI, Anthropic, Google, DeepSeek, Qwen, Doubao, Flux, Suno and Kling. One key, one bill, one set of rates — drawn live from the model catalogue.
OpenAI
GPT-6 Astra, GPT-5.6 Sol and the image model on one compatible key.
3 models · gpt-6-astra, gpt-5.6-sol, gpt-image-2-c
Anthropic
Claude Opus 5, Sonnet 4.5 and Fable 5.1 at discounted per-token rates.
3 models · claude-opus-5, claude-sonnet-4-5-20250929, claude-fable-5-1
Gemini text, Pro and Flash image models behind one OpenAI-style key.
5 models · gemini-3.1-pro-preview, gemini-3.8-flash, gemini-3-flash-preview…
DeepSeek
Frontier-level reasoning at $0.0613 per million tokens.
1 model · deepseek-v4
Qwen
Chinese-first chat and coding at $0.0784 per million tokens.
1 model · qwen3-max
Doubao
Natural low-latency speech synthesis at $0.014 per 1K characters.
1 model · doubao-voice-2
Flux
Photoreal image generation and editing at $0.021 per request.
1 model · flux-2-pro
Suno
Full songs and instrumentals generated per request.
1 model · suno-v6
Kling
Text-to-video and image-to-video shots with camera control.
1 model · kling-3.0
Why one key for every provider
No vendor sprawl
Instead of separate accounts, keys and invoices per vendor, a single OpenLux key reaches every provider in the catalogue.
Switch by model name
All providers speak the same OpenAI-compatible protocol, so moving from a GPT tier to Claude, Gemini or DeepSeek is a one-line change.
Compare on real traffic
Because every provider is metered the same way, A/B testing two vendors on your own workload costs only the tokens involved.
Pick a provider and start building
Every provider page lists live rates, the models in the catalogue and the quickstart you need. Create a key and send your first request in minutes.