CrossModel
Back to model catalog

Compare models

Context, limits, capabilities, price and use cases side by side — up to 4 models at a time. The URL holds your selection, so you can share it.

GPT-5.4GPT-5.5GPT-5.4 MiniChat with these 3
AttributeGPT-5.4openai/gpt-5.4GPT-5.5openai/gpt-5.5GPT-5.4 Miniopenai/gpt-5.4-mini
Specs
Context window1M1M400K
Max output128K128K128K
Input modalitiesTextImageTextImageTextImage
Output modalitiesTextTextText
Capabilities
StreamingSupportedSupportedSupported
Tool callingSupportedSupportedSupported
Structured output (JSON)SupportedSupportedSupported
Image inputSupportedSupportedSupported
Thinking modeSupportedSupportedSupported
Price / 1M tokens
Input$ 2.50>= 272K$ 5.00$ 5.00>= 272K$ 10.00$ 0.75
Output$ 15.00>= 272K$ 22.50$ 30.00>= 272K$ 45.00$ 4.50
Cache read$ 0.25>= 272K$ 0.50$ 0.50>= 272K$ 1.00$ 0.075
Cache write$ 2.50>= 272K$ 5.00$ 5.00>= 272K$ 10.00$ 0.75
What it is good at
Best for
  • Agentic coding
  • Professional knowledge work
  • Agentic coding
  • Professional knowledge work
  • Fast agentic coding
  • High-volume subagents
Published benchmarks
Terminal-Bench 2.0pass_rate
  • 75.1%effort xhigh(setup not fully disclosed)
  • 82.7%effort xhigh(setup not fully disclosed)
  • 60%effort xhigh(setup not fully disclosed)
BrowseComp (with tools)accuracy
  • 82.7%effort xhigh(setup not fully disclosed)
  • 84.4%effort xhigh(setup not fully disclosed)
—
SWE-Bench Pro (Public)pass_rate
  • 57.7%effort xhigh(setup not fully disclosed)
  • 58.6%effort xhigh(setup not fully disclosed)
  • 54.4%effort xhigh(setup not fully disclosed)
OSWorld-Verifiedsuccess_rate
  • 75%effort xhigh(setup not fully disclosed)
  • 78.7%effort xhigh(setup not fully disclosed)
  • 72.1%effort xhigh(setup not fully disclosed)
GDPval (wins or ties)wins_or_ties—
  • 84.9%effort xhigh(setup not fully disclosed)
—
OpenAI MRCR v2 8-needle 128K–256Kaccuracy——
  • 33.6%effort xhigh(setup not fully disclosed)
OpenAI MRCR v2 8-needle 512K–1Maccuracy—
  • 74%effort xhigh(setup not fully disclosed)
—
MMMU Pro (no tools)accuracy
  • 81.2%effort xhigh(setup not fully disclosed)
—
  • 76.6%effort xhigh(setup not fully disclosed)
MMMU Pro (with tools)accuracy—
  • 83.2%effort xhigh(setup not fully disclosed)
—
Humanity's Last Exam Full (with tools)accuracy
  • 52.1%effort xhigh(setup not fully disclosed)
  • 52.2%effort xhigh(setup not fully disclosed)
  • 41.5%effort xhigh(setup not fully disclosed)
GPQA Diamondaccuracy
  • 92.8%effort xhigh(setup not fully disclosed)
  • 93.6%effort xhigh(setup not fully disclosed)
  • 88%effort xhigh(setup not fully disclosed)
Toolathlonscore
  • 54.6%effort xhigh(setup not fully disclosed)
  • 55.6%effort xhigh(setup not fully disclosed)
  • 42.9%effort xhigh(setup not fully disclosed)