Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works
How smart is GPT-6 Luna?
GPT-6 Luna is a model from OpenAI, released on Sep 22, 2026. Its SI-IQ is 130 (90% interval 127–133), estimated from 8 benchmark results by 2 independent evaluators. Its strongest area is Math and its weakest is Reasoning.
Facts
- Developer
- OpenAI
- Released
- Sep 22, 2026
- Access
- API access
- Benchmarks used
- 8 results from 2 evaluators
- API price
- $0.10 input / $0.50 output per 1M tokens (developer's price list, Oct 18, 2026)
- Context window
- 922,000 tokens
Ability by area (SI-IQ, 90% interval)
Every benchmark result we used
Each score is one value exactly as the source published it: for Epoch AI the best of its run settings, for OpenRouter the median hosting provider (the developer's own hosting when it ran the test). “Implied SI-IQ” is the ability that this single result points to; the overall SI-IQ is their precision-weighted average.
| Benchmark | Score | Evaluated by | Source |
|---|---|---|---|
| Chess Puzzles | 36.0% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| Furniture Assembly not in index | 44.2% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| Mystery Game Puzzles | 7.0% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| FrontierMath Tier 4 | 56.1% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| FrontierMath Tiers 1–3 | 78.9% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| OTIS Mock AIME 2024–2025 | 98.9% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| GPQA Diamond | 90.5% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| GPQA Diamond (OpenRouter run) | 88.2% | OpenRouter | OpenRouter benchmark datasets · data file |
| SimpleQA Verified | 41.4% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| τ²-bench Airline (OpenRouter run) | 68.0% | OpenRouter | OpenRouter benchmark datasets · data file |
Models with similar SI-IQ
| Model | SI-IQ | |
|---|---|---|
| GPT-5.6 LunaOpenAI | 130 | GPT-6 Luna vs GPT-5.6 Luna |
| Gemini 3.8 FlashGoogle DeepMind | 130 | GPT-6 Luna vs Gemini 3.8 Flash |
| GPT-5.4OpenAI | 131 | GPT-6 Luna vs GPT-5.4 |
| Claude Haiku 5.5Anthropic | 129 | GPT-6 Luna vs Claude Haiku 5.5 |
| Gemini 3.7 FlashGoogle DeepMind | 131 | GPT-6 Luna vs Gemini 3.7 Flash |
Computed with SI-IQ method version 1.0.0. Read the full methodology
Data sources: Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0); API list prices via LiteLLM's price table (MIT); each price links to the developer's pricing page; Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter..