OpenAI: GPT-5.5

GPT-5.5 is a frontier text model from OpenAI, positioned for complex professional workloads. It builds on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It is a text model in the sense that its output is text, though it accepts more than text as input. The knowledge cutoff is 2025-12-01, and the provider release date is 2026-04-24.

On the input side, GPT-5.5 accepts text, images, and files, and it produces text output. The context window is 1,050,000 tokens, which is large enough to hold very long documents, codebases, or multi-file inputs in a single request. The maximum output is 128,000 tokens, so long generations are possible within one call. Supported parameters include reasoning controls (reasoning, reasoning_effort, include_reasoning, verbosity), output shaping options (response_format, structured_outputs), sampling and determinism controls (seed), and tool use (tools, tool_choice), along with max_tokens and max_completion_tokens for output length.

Those capabilities point to a few practical uses. The large context window and file input suit analysis of long contracts, research papers, or log dumps where the whole document needs to be in view at once. Image input allows tasks that mix screenshots, diagrams, or scanned pages with text instructions. Reasoning effort and verbosity controls let you trade latency and output detail against task difficulty, which is useful for agent-style workflows where the model calls tools and needs to decide when to stop. Structured outputs and response_format help when the result must be parsed by code rather than read by a person, and seed helps when you want repeatable runs during testing.

You call GPT-5.5 through api14's OpenAI-compatible API using the model id "openai/gpt-5.5". Because the interface follows the OpenAI chat completions shape, existing OpenAI SDKs and clients work with only a base URL and key change. Requests can set any of the supported parameters listed above: reasoning, reasoning_effort, include_reasoning, verbosity, max_tokens, max_completion_tokens, response_format, structured_outputs, seed, tools, and tool_choice. Note that max_tokens and max_completion_tokens both control output length, so pick one convention and stay consistent.

There is no stated limitation in the facts beyond the output cap of 128,000 tokens and the knowledge cutoff of 2025-12-01, so anything after that date is outside the model's training. If your task is simple text generation where reasoning controls and tool use add no value, a smaller model may be cheaper and faster, but the facts here do not name one. For work that needs image or file input, long context, or structured output, GPT-5.5 covers those cases directly.