Most factually accurate LLM
- Gemini 4 Argon leads
- 76.0%
- models ranked
- 77
- lead over #2
- +0.4 pts
SimpleQA Verified · read October 22
As of October 22, 2026, Google Gemini 4 Argon leads 77 models on SimpleQA Verified with 76.0%, ahead of OpenAI GPT-6 Astra (75.6%) and OpenAI GPT-6.1 Sol (73.9%).
The full best LLM for knowledge ranking combines this benchmark with others.
| Rank | Model | SimpleQA Verified | Coding evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Gemini 4 Argon Google | 76.0% | 81.0 | $10.00 | |
| 2 | GPT-6 Astra OpenAI | 75.6% | 85.1 | $50.00 | |
| 3 | GPT-6.1 Sol OpenAI | 73.9% | 82.4 | $10.00 | |
| 4 | Gemini 3.1 Pro Preview Google | 73.5% | 32.3 | $12.00 | |
| 5 | Claude Opus 5.5 Anthropic | 72.2% | 88.8 | $20.00 | |
| 6 | Claude Fable 5.1 Anthropic | 70.8% | 84.2 | $50.00 | |
| 7 | Claude Fable 5 Anthropic | 70.7% | 79.3 | $50.00 | |
| 8 | Gemini 3.8 Flash Google | 69.7% | 61.6 | $3.75 | |
| 9 | GPT-5.6 Sol OpenAI | 69.7% | 73.5 | $20.00 | |
| 10 | Gemini 3.7 Flash Google | 69.2% | 58.0 | $3.75 | |
| 11 | Gemini 3 Flash Preview Google | 66.8% | 43.7 | $3.00 | |
| 12 | Gemini 3.5 Flash Google | 66.2% | 39.8 | $9.00 | |
| 13 | Gemini 3.6 Flash Google | 66.2% | 62.5 | $3.75 | |
| 14 | GPT-5.5 OpenAI | 63.0% | 71.3 | $30.00 | |
| 15 | GPT-6 Sol OpenAI | 60.7% | 76.0 | $10.00 | |
| 16 | Muse Spark 1.2 Meta | 60.3% | 55.7 | $4.25 | |
| 17 | Claude Opus 5 Anthropic | 59.9% | 83.0 | $25.00 | |
| 18 | Muse Spark 1.1 Meta | 57.8% | 70.3 | $4.25 | |
| 19 | Grok 4.7 xAI | 56.0% | 66.7 | $6.00 | |
| 20 | Qwen3.7 Max Qwen | 55.8% | 54.1 | $4.425 |
How this list is built
Generated from the dataset on every refresh. Nobody picks the order by hand.
Models by their SimpleQA Verified result, as the board publishes it, best first. Short factual questions with one checkable answer; it rewards knowing, and not guessing. Share answered correctly.
- OpenRouter, fetched
- Epoch AI benchmarking hub, fetched