deepseek-v4.1-flash
DeepSeek
Claude, GPT, Gemini, DeepSeek and 29 leading models behind one endpoint, billed by usage,
on the Shared, Dedicated or Enterprise channel.


Models from OpenAI, Anthropic, Google, DeepSeek and 8 providers in total, on one key and one bill.
Chat, coding and image generation share one key and one bill, and new models are callable on launch day.
The full GPT Image and Nano Banana lineups, with image-to-image and inpainting, billed per image or per token.
Minimalist concrete architecture with a curved staircase, black-and-white film look
gpt-image-2
Portrait of a woman on a rainy street at night, cinematic black-and-white photography
gemini-3-pro-image
Glass perfume bottle on a stone pedestal, side window light, luxury advertising style
gpt-image-2.5-sunburst
A little robot reading by the window in soft morning light
gemini-3.1-flash-image
Turn the second image into color, keep the composition unchanged
gpt-image-2.5-flare
Generate 4 product hero shots from different angles
gemini-2.5-flash-image

Switch between Claude, GPT and Gemini by changing one model name — no new keys, no code changes.
Summarize this API doc into three key points
claude-sonnet-5-5
Done: authentication, rate limits, and error codes.
Switch to another model and write an English version
gpt-6-sol
Done. Three key points, same structure.
Add a product cover image
gpt-image-2
Generated 1 image at 1024×1024.
Every key is bound to a channel, so different jobs can use different channels.
OpenAI, Anthropic and Gemini protocols work as-is; coding tools and frameworks only need a new base URL.
Compatible with official SDKs; migration only takes a new base URL and key.
Streaming responses are relayed chunk by chunk instead of being buffered in full.
No subscription and no monthly fee — usage-based charges with exportable statements.
Shared, Dedicated and Enterprise channels with different capacity and pricing.
Requests automatically retry on backup channels when an upstream fails.
Organizations top up centrally and assign quotas per member.
Every model's price in each group is published on the pricing page.
Invoices for the Dedicated and Enterprise channels (not VAT invoices).
New models are added as soon as they launch, with prices published alongside.
deepseek-v4.1-flash
DeepSeek
GPT-5.6 Terra
OpenAI · 1.05M
grok-4.6
claude-fable-5
Anthropic
Claude Opus 4.8
Anthropic · 1M
Claude Sonnet 5
Anthropic · 1M
gemini-3.1-pro
gemini-3.6-flash
GPT-5.5
OpenAI · 1M
gpt-6.1-sol
OpenAI
GPT-5.6 Luna
OpenAI · 1.05M
GPT-6 Astra
OpenAI · 1.05M
grok-4.7
Claude Fable 5.1
Anthropic · 1M
Claude Opus 5
Anthropic · 1M
Claude Sonnet 5.5
Anthropic · 1M
Gemini 3.5 Flash
Google · 1M
gemini-3.7-flash
GPT-6 Luna
OpenAI · 1.05M
gemini-4-argon
GPT-5.6 Sol
OpenAI · 1.05M
grok-4.5
grok-build-0.1
Claude Haiku 4.5
Anthropic · 200K
claude-opus-5-5
Anthropic
Gemini 3.1 Flash Lite
Google · 1M
gemini-3.5-flash-lite
gemini-3.8-flash
GPT-6 Sol
OpenAI · 1.05M
Personal and organization accounts are both welcome. Pay as you go and switch channels anytime.