OpenAI
GPT-5.6 Terra (Non-reasoning)
Quality Score40.8
Value Score28.5
Reliability38.4
Cache Discount90%
Model specification
- Reasoning
- non-reasoning
- Input modalities
- text, image
- Output modalities
- text
- Context window
- 1M tokens
- Weights
- closed weights
Token prices USD per 1M tokens
- Input
- $2.00
- Output
- $12.00
- Cached Input
- $0.200
- Cached Output
- $2.50
- Cost per task
- $0.103
Capability
- Intelligence
- 34.6
- Coding
- 52.3
- Agentic
- 30.1
- Omniscience
- -23.2
- Correct
- 36.8%
- Blended price
- $1.74
Answer outcomes
Correct 36.8%Incorrect 60.0%Abstained 3.1%
Artificial Analysis benchmarks
- GDPval-AA v2
- 37%
- τ³-Banking
- 16%
- Terminal-Bench v2.1
- 56%
- SciCode
- 45%
- Humanity’s Last Exam
- 11%
- GPQA Diamond
- 75%
- CritPt
- 2%
- AA-Omniscience
- 38%
- AA-LCR
- 55%
Similar models
- GPT-5.6 Luna (Non-reasoning)OpenAIQuality Score 32.2Value Score 29.1
- GPT-5.6 Sol (Non-reasoning)OpenAIQuality Score 56.7Value Score 32.3
- GPT-5.6 Luna (low)OpenAIQuality Score 42.2Value Score 37.7
- o3OpenAIQuality Score 38.8Value Score 27.3
- GPT-5.5 Instant (June 2026)OpenAIQuality Score 44.5Value Score 24.8
- GPT-5.6 Luna (medium)OpenAIQuality Score 46.4Value Score 42.0