GPT-5.5

LLM Provider / Model · Frontier reasoning, agentic coding, long-context refactors, multimodal analysis, replaces GPT-5.4 as default flagship

At a glance

Input price	$5
Output price	$30
Cache price	$0.50
Price tier	Premium
Context window	1M+
Max output	128K
Context tier	500K+
Speed tier	Standard
Latency	medium
Knowledge cutoff	Jan 2026
Modality	Multimodal (vision)
Model ID	gpt-5-5
Provider	OpenAI
HumanEval	95%
MMLU	89%
SWE-Bench	75%
Benchmark	Terminal-Bench 2.0: top
Released	Apr 2026
Hosting	Closed/API
Capabilities	Vision, Tool use, Streaming, Structured output, Extended reasoning, Function calling, Prompt caching, Agentic workflows

Vision, Tool use, Streaming, Structured output, Extended reasoning, Function calling, Prompt caching, Agentic workflows

Frontier reasoning, agentic coding, long-context refactors, multimodal analysis, replaces GPT-5.4 as default flagship

Kiro's spec mode is tuned for Claude models — GPT-5.5 may work but spec-driven flows are best with Claude.
Gemini CLI is locked to Google Gemini models — your LLM is GPT-5.5.
Claude Code only works with Claude models — your LLM is GPT-5.5.
Devin uses Cognition's proprietary model internally — your selected GPT-5.5 will be ignored.
OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected GPT-5.5 will be ignored.
Jules runs Gemini models internally — your selected GPT-5.5 will be ignored.
The Claude Code SDK is tuned for Claude models — GPT-5.5 may work via the model adapter but loses Claude-specific features (thinking, prompt caching, sub-agents).
Lovable uses a fixed internal model — your selected GPT-5.5 will be ignored.
Replit Agent 3 uses its own model selection internally — your selected GPT-5.5 will be ignored.

Build a full stack around GPT-5.5 — Flowpicker shows compatibility warnings before you commit.