Models
OpenAI text-to-speech model for natural-sounding voice generation.
OpenAI Whisper automatic speech recognition and transcription model.
OpenAI DALL·E 2 image generation model.
OpenAI DALL·E 3 text-to-image generation model.
OpenAI updated DALL·E 3 image generation model.
OpenAI GPT-3.5 Turbo model for fast, low-cost chat and text workloads.
OpenAI GPT-4.1 mini model for long-context multimodal workloads with lower cost than the full GPT-4.1 tier.
OpenAI GPT-4.1 nano model optimized for very low-cost, low-latency long-context tasks.
OpenAI multimodal flagship model for text, vision, audio, and general-purpose API workloads.
OpenAI cost-efficient multimodal model for high-volume text, vision, and audio workloads.
OpenAI GPT-4 Turbo model for high-capability text and vision tasks with a 128K context window.
OpenAI previous frontier GPT-5 model for professional work with configurable reasoning effort.
OpenAI GPT-5.3 Instant model used in ChatGPT.
OpenAI agentic coding model optimized for Codex-style software engineering tasks.
OpenAI GPT-5.4 model for complex professional work.
OpenAI compact GPT-5.4-class model for high-volume coding, computer use, and subagent workloads.
OpenAI lowest-cost GPT-5.4-class model for simple high-volume tasks.
OpenAI high-compute GPT-5.4 variant for more precise responses.
OpenAI frontier model for complex coding and professional work.
OpenAI high-compute GPT-5.5 variant for difficult problems and background-mode workflows.
OpenAI GPT-5 mini model optimized for cheaper high-volume GPT-5 family workloads.
OpenAI realtime model for low-latency text and audio interactions.
OpenAI o1 reasoning model for complex math, coding, and science problems.
OpenAI reasoning model for complex analysis, coding, math, and tool-heavy workflows.
OpenAI compact reasoning model designed for cost-efficient reasoning and coding workloads.
OpenAI compact reasoning model with multimodal input support for efficient reasoning workloads.
OpenAI Sora text-to-video generation model.
OpenAI Sora 2 text-to-video generation model with synchronized audio.