Ctrl K
Back to models

OpenAI

GPT-5.6 Terra

  • Released
  • Closed weights
Compare
Intelligence Index

Max setting

42.1#17 of 83measured models
Measured · Artificial Analysis v4.3.2
API price from
$2.00 in / $12.00 out / 1M tokOpenAI · direct API
Context window
1.05M tok

Performance profile

ReasoningRank #11 of 107
92.5%
GPQA Diamond
CodingRank #18 of 94
55.0%
SciCode
Cost efficiencyRank #51 of 67
30.1pts / $
Intelligence ÷ eval cost

Capabilities

Context window

Modalities

Reasoning modes

Avg reasoning tokens / task — Artificial Analysis Intelligence Index v4.3.2: weighted average reasoning tokens per benchmark task, not a token limit. Source (opens in a new tab)

Integration

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
GPT-5.6 Terra
Cost per task (USD) · log scale
GPT-5.6 TerraMax
Intelligence Index
42.1
Cost / task
$1.40
Evaluation time
4.7m

Pricing & access

API route
OpenAIAPI
Tier & context

USD per 1M tokens

Input
$2.00
Output
$12.00
Cached input
$0.20
Responses API tool compatibility (opens in a new tab)
  • Web search
  • File search
  • Image generation
  • Code execution
  • Shell
  • File edits
  • Skills
  • Computer use
  • MCP
  • Tool search

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

GPT-5.6 Terra · Max#20 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 17JT-4.1 Flash 236B A21BReasoning43.6%
  6. 18Grok 4.7Extra high43.1%
  7. 19Qwen3.8 Max (0902)Reasoning43.1%
  8. 20GPT-5.6 TerraMax42.9%
  9. 21GPT-5.3-CodexExtra high42.5%
  10. 22Qwen3.8 2.4T A95BReasoning42.4%
  11. 23GLM-5.3Max42.3%

Peers use their best measured result on this benchmark.

Alternatives

Each model at its highest-scoring measured setting · Intelligence Index v4.3.2 and its evaluation cost per task.