OpenAI
GPT-5.5 is an OpenAI API model described for coding, professional work, complex production workflows, and tool-heavy use cases, with text/image input, text output, configurable reasoning effort, and token-based API pricing.
OpenAI GPT-5.5 is strongly evidenced as a frontier API model for coding, tool-heavy agents, professional work, long-context retrieval, and grounded workflows. Official OpenAI sources describe text and image input, text output, configurable reasoning effort, pricing, developer guidance, and safety evaluation under the Preparedness Framework. These sources support high technical, developer-experience, and multimodal scores, though the provided material does not establish broad output modalities beyond text. Coding evidence is the strongest area. DeepSWE/DataCurve reports GPT-5.5 [xhigh] at rank 1 with 70% ±3% Pass@1, SWE-bench secondary sources report 88.7% on SWE-bench Verified, and Terminal-Bench lists GPT-5.5 entries above 82%. LiveCodeBench was checked, but the accessible official leaderboard did not expose a stable GPT-5.5 row, so it should not be used as a positive benchmark claim. Pricing is clear but expensive: supplied pricing is $5/M input, $0.50/M cached input, and $30/M output, with Batch and residency modifiers. Artificial Analysis reports 55.5 output tokens/sec but very high 97.87s TTFT, which pulls down speed. Evidence quality is good but not perfect because some benchmark support is secondary, some entries are variant- or harness-dependent, and the catalog metadata contains URL/title consistency warnings.
