Responsible AI Use Disclaimer: The tools listed are for informational purposes. Users are responsible for adhering to ethical guidelines. Learn more.

OpenAI ProprietaryMar 2026

GPT-5.5

OpenAI’s 2026 flagship — strongest at reasoning, coding and tool use.

Intelligence index
82/ 100
vs all models95th pctile
Composite of MMLU, GPQA, MATH & HumanEval
Speed
95tok/s
vs all models50th pctile
Median across providers, steady state
Blended price
$7.50/ 1M tokens
vs all models14th pctile
3:1 input:output blend

At a glance

Context window
400k tokens
Max output
16k tokens
Input price
$5.00 / 1M tokens
Output price
$15.00 / 1M tokens
Time to first token
0.42s
Input modalities
text, image
Output modalities
text
License
Proprietary
Provider
OpenAI

Benchmark scores

Public scores from each provider; bars compare this model against the leader in each benchmark.

MMLU
General knowledge across 57 subjects
90.2
leader: 91.8
MMLU Pro
Harder MMLU successor with more reasoning
78.0
leader: 80.0
GPQA
Graduate-level science Q&A
62.5
leader: 78.0
MATH
Competition mathematics
89.1
leader: 94.8
HumanEval
Python code generation pass@1
93.0
leader: 95.8
Strengths
  • Best-in-class reasoning
  • Huge 400k context
  • Strong tool use and agents
Weaknesses
  • Expensive vs Sonnet for non-reasoning tasks
  • Higher latency than gpt-5.5-mini
Best for
  • Agentic workflows
  • Complex coding
  • Hard math & research

GPT-5.5 vs… popular head-to-heads

One-click matchups against the models people compare GPT-5.5 with most.

GPT-5.5 — frequently asked questions

GPT-5.5 is a large language model from OpenAI, released on 15 March 2026. OpenAI’s 2026 flagship — strongest at reasoning, coding and tool use.

Need help choosing between models?

Compare every option in one sortable table — intelligence, speed and price on a single page.