Inference

OpenAI-compatible LLM API with access to chat, transcription, and embedding models

We provide access to pre-configured LLM models through our inference service. This allows developers and researchers to focus on building applications and developing new AI solutions without managing models, infrastructure, or supporting systems. The service provides access to modern language models through an API compatible with the standard OpenAI API specification, enabling straightforward integration with existing applications and development tools while reducing implementation effort.

Available models

• qwen3.5-122b — for complex reasoning, math, coding, analysis; thinking enabled
• qwen3.5-122b-nonthinking — for fast responses, Q&A, translation, summarization; thinking disabled
• qwen3.6-35b-a3b — for fast reasoning: coding, math, structured problem solving, and technical questions
• gemma-4-31B-it — for general-purpose tasks, instruction following; multilingual
• gemma-4-26b-a4b — for general-purpose tasks, instruction following, general productivity with a speed/quality trade-off
• whisper-large-v3 — for audio transcription, multilingual speech-to-text
• qwen3-embedding-8B — for text retrieval, semantic search, RAG pipelines, multilingual classification

If you would like to get more information about the service or would like to order it for your project, please contact us!