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-6 LunaGPT-6 SolGPT-5.6 LunaChat with these 3
AttributeGPT-6 Lunaopenai/gpt-6-lunaGPT-6 Solopenai/gpt-6-solGPT-5.6 Lunaopenai/gpt-5.6-luna
Specs
Context window1M1M1M
Max output128K128K128K
Input modalitiesTextImageTextImageTextImage
Output modalitiesTextTextText
Capabilities
StreamingSupportedSupportedSupported
Tool callingSupportedSupportedSupported
Structured output (JSON)SupportedSupportedSupported
Image inputSupportedSupportedSupported
Thinking modeSupportedSupportedSupported
Price / 1M tokens
Input$ 0.10>= 272K$ 0.20$ 2.00>= 272K$ 4.00$ 0.20>= 272K$ 0.40
Output$ 0.50>= 272K$ 0.75$ 10.00>= 272K$ 15.00$ 1.20>= 272K$ 1.80
Cache read$ 0.010>= 272K$ 0.020$ 0.20>= 272K$ 0.40$ 0.020>= 272K$ 0.040
Cache write$ 0.13>= 272K$ 0.25$ 2.50>= 272K$ 5.00$ 0.25>= 272K$ 0.50
What it is good at
Best for
  • High-volume tasks
  • Everyday coding
  • Agentic coding
  • Professional workflows
  • High-volume agent execution
  • Routine implementation and subagents
Published benchmarks
Terminal-Bench 2.1pass_rate——
  • 84.7%effort max(setup not fully disclosed)
BrowseComp (with tools)accuracy——
  • 83.3%effort max(setup not fully disclosed)
Agents' Last Examscore—
  • 56.4%effort max(setup not fully disclosed)
  • 50.3%effort max(setup not fully disclosed)
DeepSWE 1.1resolved_rate
  • 66.6%effort max(setup not fully disclosed)
  • 68.8%effort max(setup not fully disclosed)
—
SWE-Bench Pro (Public)pass_rate——
  • 62.7%effort max(setup not fully disclosed)
OSWorld 2.0success_rate——
  • 45.6%effort max(setup not fully disclosed)
OSWorld 2.0 (v2026.08.08 offline, partial score)score—
  • 60.5%effort xhigh(setup not fully disclosed)
—
OpenAI MRCR v2 8-needle 512K–1Maccuracy——
  • 41.3%effort max(setup not fully disclosed)
MMMU Pro (with tools)accuracy——
  • 79.5%effort max(setup not fully disclosed)
GPQA Diamondaccuracy——
  • 92.3%effort max(setup not fully disclosed)
Toolathlonscore——
  • 53.4%effort max(setup not fully disclosed)