GPT-5.6 Luna
Jul 2026 · Proprietary · Current · estimated evidence
The punchline of the 5.6 stack: near-Terra coding at Luna prices. Default OpenAI pick for volume.
Intelligence
52.3
AA Index
API price
$0.20 / $1.20
Input / output per 1M
Context
1.05M
202 tok/s
Capability
Intelligence Index52.3
Coding index71.4
Agentic index46.9
Arena Elo (offset)—
Use it when
- ▸High volume
- ▸Customer support
- ▸Cheap coding loops
Skip if
- –You need Opus-class agents or HLE
Benchmarks
GPQA Diamond Graduate-level science questions designed so Google search is not enough. Still one of the cleanest knowledge/reasoning splits. | 92.3% |
|---|---|
Humanity’s Last Exam Expert-written questions across fields. Harder than MMLU; the current differentiator for “does this model actually know things.” | — |
MMLU-Pro Harder, less-saturated successor to MMLU. Classic MMLU is above 90% for every flagship and no longer ranks the field. | — |
SWE-bench Verified 500 human-validated GitHub issues. Score swings 5–15 points by harness — treat vendor numbers as an upper bound. | — |
SWE-bench Pro Harder, contamination-resistant coding eval. Currently the best public split between “can code” and “can maintain a repo.” | 62.7% |
Terminal-Bench 2.1 End-to-end tasks in a real terminal. Better proxy for coding agents than HumanEval, which is fully saturated. | 84.7% |
ARC-AGI-2 Abstract visual puzzles. Rewards generalization over memorization. GPT-5.6 Sol currently leads the published set. | — |
AIME 2025 American Invitational Mathematics Examination. Contest math; reasoning-mode models dominate. | — |
MMMU College-level multimodal understanding across diagrams, charts, and exam figures. | — |
Modalities
text · vision · tools
Reasoning mode
reasoning
Hybrid and reasoning models spend tokens thinking. That raises GPQA and agents, and also raises latency and bill.