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-6.1 SolGPT-6 AstraGPT-6 LunaGPT-6 SolChat with these 4
AttributeGPT-6.1 Solopenai/gpt-6.1-solGPT-6 Astraopenai/gpt-6-astraGPT-6 Lunaopenai/gpt-6-lunaGPT-6 Solopenai/gpt-6-sol
Specs
Context window1M1M1M1M
Max output128K128K128K128K
Input modalitiesTextImageTextImageTextImageTextImage
Output modalitiesTextTextTextText
Capabilities
StreamingSupportedSupportedSupportedSupported
Tool callingSupportedSupportedSupportedSupported
Structured output (JSON)SupportedSupportedSupportedSupported
Image inputSupportedSupportedSupportedSupported
Thinking modeSupportedSupportedSupportedSupported
Price / 1M tokens
Input$ 2.00>= 272K$ 4.00$ 10.00>= 272K$ 20.00$ 0.10>= 272K$ 0.20$ 2.00>= 272K$ 4.00
Output$ 10.00>= 272K$ 15.00$ 50.00>= 272K$ 75.00$ 0.50>= 272K$ 0.75$ 10.00>= 272K$ 15.00
Cache read$ 0.10>= 272K$ 0.20$ 1.00>= 272K$ 2.00$ 0.010>= 272K$ 0.020$ 0.20>= 272K$ 0.40
Cache write$ 2.50>= 272K$ 5.00$ 12.50>= 272K$ 25.00$ 0.13>= 272K$ 0.25$ 2.50>= 272K$ 5.00
What it is good at
Best for
  • Complex coding
  • Computer use
  • Professional work
  • Software engineering
  • Professional deliverables
  • Research with computer use
  • Scientific reasoning
  • High-volume tasks
  • Everyday coding
  • Agentic coding
  • Professional workflows
Published benchmarks
Terminal-Bench 4.0pass_rate—
  • 57.9%(setup not fully disclosed)
——
BrowseComp (with tools)accuracy—
  • 91.5%(setup not fully disclosed)
——
Agents' Last Examscoresetups differ by: reasoning—
  • 59.3%(setup not fully disclosed)
—
  • 56.4%effort max(setup not fully disclosed)
AutomationBench v1.0.6score
  • 36.1%effort max(setup not fully disclosed)
———
DeepSWE 1.1resolved_ratesetups differ by: reasoning
  • 75.2%effort high(setup not fully disclosed)
  • 74.1%(setup not fully disclosed)
  • 66.6%effort max(setup not fully disclosed)
  • 68.8%effort max(setup not fully disclosed)
OSWorld 2.0 (v2026.08.08 offline, partial score)scoresetups differ by: reasoning
  • 71.4%effort max(setup not fully disclosed)
  • 72.6%(setup not fully disclosed)
—
  • 60.5%effort xhigh(setup not fully disclosed)
ExploitBenchscore
  • 99.7%effort max(setup not fully disclosed)
———
GDP.pdfscore
  • 32%effort high(setup not fully disclosed)
———
HealthBench Professionalscore
  • 64.2%(setup not fully disclosed)
———
OpenAI MRCR v2 8-needle 512K–1Maccuracy—
  • 96.3%(setup not fully disclosed)
——
FrontierMath Tier 4 (v2)score—
  • 97.6%(setup not fully disclosed)
——
GPQA Diamondaccuracy—
  • 96%(setup not fully disclosed)
——
Terminal-Bench-Science 0.1accuracy
  • 57%effort max(setup not fully disclosed)
———