Model catalog

Fast, economical GPT-5.6 model for high-volume apps, quick answers, and lightweight agent work.

Balanced GPT-5.6 model for everyday reasoning, coding, writing, and tool-enabled workflows.

Flagship GPT-5.6 model for demanding reasoning, complex production work, and high-confidence synthesis.

Lowest-cost GPT tier for short utility prompts, classification, routing, and lightweight app responses.

Strong mini GPT model for coding, computer use, and subagent-style work.

Coding-focused GPT model for repository edits, debugging, reviews, and terminal or IDE workflows.

General GPT model for everyday reasoning, coding help, writing, and tool use.

Higher-cost GPT model for advanced research, careful comparisons, and difficult synthesis.

Fast multilingual text model for quick chat, lightweight app responses, and everyday requests.

Cloudflare Workers AI model tuned for agentic coding, repository edits, and tool-heavy engineering work.

Large-context Cloudflare Workers AI text model for reasoning, writing, and longer general-purpose tasks.

Code-specialized Cloudflare Workers AI model for debugging, refactors, and end-to-end coding workflows.

Instruction-tuned Gemma 4 text model exposed through Cloudflare Workers AI for general chat and writing tasks.

Proprietary Imole transcription service exposed through the Imole API for multilingual speech-to-text, including African languages such as Fon, Adja, and Ewe.

Azure Speech transcription for files, captions, and cost-sensitive audio analysis.

Speech-to-text for transcripts, subtitles, diarization, and audio analysis.

Image generation model for prototypes, product visuals, and creative workflows.

Higher quality image generation model for production visual work.

Proprietary Imole premium TTS service exposed through the Imole API for natural narration, high-fidelity voiceovers, and reference-guided speech.

Proprietary Imole standard TTS service exposed through the Imole API for styled narration, generated voices, and reference-based voice cloning.

Proprietary Imole locale TTS service exposed through the Imole API for lightweight voice output in languages like Fon, Ewe, Kirundi, Malagasy, and Bambara.

Proprietary Imole lightweight TTS service exposed through the Imole API for fast speech generation without heavy GPU models.

Azure Speech text-to-speech for narration and voice output with selectable voices.

OpenAI text-to-speech for app voices, narration, and product audio.

Natural text-to-speech for voiceovers, narration, product audio, and apps.

High-quality multilingual text-to-speech for long-form and polished voice work.

Low-latency multilingual text-to-speech for interactive voice experiences.

Balanced speed and quality for multilingual text-to-speech.

Design a new voice from a prompt for custom voice workflows.

Multilingual speech-to-speech voice conversion from uploaded audio.

English speech-to-speech voice conversion from uploaded audio.

Translate and dub uploaded audio or video into another language.

Music generation for songs, instrumentals, background tracks, and jingles.

Sound effects and background audio from text prompts.

Remove noise and isolate speech or vocals from uploaded audio.

Sora 2 video generation billed per second with 4, 8, or 12 second duration controls.

Proprietary Imole video generation service exposed through the Imole API for text-to-video, image-to-video, and first-last-frame motion workflows.

Proprietary Imole parallax video service exposed through the Imole API for turning one image into short 3D-style camera motion clips.