GPT-6.2 Astra and GPT-6.2 Sol are now live. Try them now

Model Directory

Find your model.

Every request routed through EU infrastructure. Compare specs, check pricing, pick what works.

PopularDeepSeek

DeepSeek V4 Pro 0813

DeepSeek V4 Pro 0813 is the official release of DeepSeek V4 Pro, superseding the April preview. A 1.6T-parameter sparse mixture-of-experts text model with a 1M-token context window, function calling, and optional reasoning, with greatly enhanced agentic capabilities and a DSpark speculative decoding module.

Input: USD 2 / 1M tokensOutput: USD 4 / 1M tokensCached input: USD 0.5 / 1M tokens
PopularZhipu AI

GLM 5.3 Flash

GLM-5.3-Flash is Z.ai's natively multimodal 320B-A18B mixture-of-experts model for reasoning, coding, agentic work, long-context tasks, images, and video.

Input: USD 0.1 / 1M tokensOutput: USD 0.35 / 1M tokensCached input: USD 0.0175 / 1M tokens
PopularDeepSeek

DeepSeek V4.1 Flash

DeepSeek V4.1 Flash is an open-weight multimodal Mixture-of-Experts reasoning model with native text and image input, text output, a one-million-token context window, and a focus on agentic and coding workloads.

Input: EUR 0.2 / 1M tokensOutput: EUR 1 / 1M tokensCached input: EUR 0.01 / 1M tokens

Loading models…

One endpoint, 164 models.

EUrouter routes your requests to GPT-5, Claude, Gemini, Mistral, and dozens more — all through a single OpenAI-compatible API. Every request stays in the EU. Prices use the units shown for each model. Switch models with a single parameter change, no code rewrites needed.