Best proprietary models in 2026
The strongest proprietary models in the index, ranked by average benchmark score. 229 models fall into this group.
What the data shows
GPT-5.2 Pro leads the proprietary group at 89.6. The group median is 79.0.
The price range runs from $1.48 to $21.39 per million input tokens, so 'best' and 'cheapest' are different shortlists.
How to read the ranking
Proprietary models are API-only in this index. That simplifies operations but limits fine-tuning and on-prem deployment.
- Check the licence before planning a deployment.
- Compare hosted and self-hosted cost if open weights are available.
- Test on your own data.
- Keep a fallback from the other licensing group.
Best proprietary models
| # | Model | Provider | Input / 1M | Output / 1M | Context | Avg score |
|---|---|---|---|---|---|---|
| 1 | GPT-5.2 Pro | OpenAI | $21.32 | $76.33 | 2,000,000 | 89.6 |
| 2 | GPT-5.2 Max | OpenAI | $21.39 | $90.69 | 1,000,000 | 89.2 |
| 3 | Claude Opus 4.5 Pro | Anthropic | $20.98 | $85.59 | 200,000 | 89.1 |
| 4 | GPT-5.1 Reasoning | OpenAI | $20.27 | $82.11 | 400,000 | 88.3 |
| 5 | Claude Opus 4.5 Reasoning | Anthropic | $20.91 | $108.13 | 2,000,000 | 88.2 |
| 6 | GPT-5.2 | OpenAI | $19.71 | $85.71 | 2,000,000 | 88.0 |
| 7 | GPT-5.1 Max | OpenAI | $20.50 | $103.98 | 2,000,000 | 87.9 |
| 8 | Gemini 3 Pro Max | Google DeepMind | $20.57 | $112.92 | 2,000,000 | 87.9 |
| 9 | Grok 4 Reasoning | xAI | $19.67 | $93.29 | 2,000,000 | 87.8 |
| 10 | GPT-5.1 Pro | OpenAI | $19.77 | $100.32 | 200,000 | 87.8 |
| 11 | Claude Opus 4.5 Max | Anthropic | $20.90 | $106.51 | 1,000,000 | 87.8 |
| 12 | Claude Sonnet 4.5 Reasoning | Anthropic | $19.48 | $68.20 | 400,000 | 87.7 |
| 13 | Claude Opus 4.5 | Anthropic | $19.42 | $75.27 | 400,000 | 87.7 |
| 14 | Gemini 3 Pro Reasoning | Google DeepMind | $19.69 | $98.36 | 200,000 | 87.5 |
| 15 | o4 Reasoning | OpenAI | $19.69 | $103.53 | 1,000,000 | 87.5 |
| 16 | GPT-5.2 Reasoning | OpenAI | $20.98 | $102.78 | 1,000,000 | 87.5 |
| 17 | GPT-5.2 Vision | OpenAI | $19.03 | $85.72 | 1,000,000 | 87.4 |
| 18 | GPT-5.2 Turbo | OpenAI | $20.73 | $96.05 | 1,000,000 | 87.4 |
| 19 | o4 Pro | OpenAI | $19.37 | $70.19 | 1,000,000 | 87.3 |
| 20 | Gemini 3 Pro Pro | Google DeepMind | $19.80 | $108.61 | 200,000 | 87.3 |
Methodology
This story uses the AI & LLM Index dataset: 600 models, 40 providers and 12 benchmarks. Prices are listed per million tokens. Scores are modelled estimates on a 0–100 scale and are not official benchmark results. Figures are rounded for readability.
Related
FAQ
What does the Best proprietary models in 2026 data show?
The strongest proprietary models in the index, ranked by average benchmark score. 229 models fall into this group.
Are the numbers official?
No. Prices and scores are compiled for this demo index. Benchmark values are modelled estimates, not official results. Verify figures with the provider or benchmark owner before making decisions.
How should I use this ranking?
Use it to form a shortlist. Check the specific model pages, compare price and context, and test the shortlist on your own examples before committing.