FACTS Search leaderboard
As of 6 Nov 2026, the top score on FACTS Search is 93.3% by GPT-6 Astra (OpenAI), ahead of GPT-6.1 Sol at 92.8%, across 15 model configurations with a published result (Google, result dated 5 Nov 2026). Using web search as a tool and synthesising what comes back without inventing anything. Published by Google. FACTS Search is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- GPT-6 Astra holds the top FACTS Search score at 93.3% (Google, 5 Nov 2026).
- GPT-6.1 Sol is second on FACTS Search at 92.8% (Google, 5 Nov 2026).
- FACTS Search: Using web search as a tool and synthesising what comes back without inventing anything. Published by Google (Google).
- 15 model configurations have a published FACTS Search result on BenchLeader (BenchLeader, 6 Nov 2026).
- FACTS Search does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Using web search as a tool and synthesising what comes back without inventing anything. Published by Google.
- Published by
- Googledata via Kaggle Benchmarks
- Category
- Agents & tools
- Index weight
- Reference only
- Models
- 15
- Data as of
- 6 Nov 2026
- Newest result
- 5 Nov 2026
Kaggle (kaggle.com); each board credits its publishing lab.
- 1GPT-6 Astra93.3%
- 2GPT-6.1 Sol92.8%
- 3Kimi K390.3%
- 4Gemini 3.1 Pro90.2%
- 5Gemini 4 Argon89.9%
- 6Claude Opus 5.588.4%
- 7GLM 5.387.7%
- 8Claude Sonnet 5.586.8%
- 9GPT-6 Luna86.8%
- 10Qwen3.8 Max (max)85.0%
- 11Gemini 3.8 Flash80.4%
- 12Claude Haiku 5.580.3%
- 13Gemini 3.5 Flash Lite69.0%
- 14Gemma 4 26B A4B66.4%
- 15gpt-oss-120b57.6%
| # | ||||
|---|---|---|---|---|
| 1 |  GPT-6 Astra | 93.3% | 68.2 | 2026-11-05 |
| 2 | GPT-6.1 Sol | 92.8% | 62.5 | 2026-11-05 |
| 3 |  Kimi K3open ↗ | 90.3% | 59.1 | 2026-11-05 |
| 4 |  Gemini 3.1 Pro | 90.2% | 60.5 | 2026-11-05 |
| 5 | Gemini 4 Argon | 89.9% | 63.4 | 2026-11-05 |
| 6 |  Claude Opus 5.5 | 88.4% | 63.4 | 2026-11-05 |
| 7 |  GLM 5.3open ↗ | 87.7% | – | 2026-11-05 |
| 8 | Claude Sonnet 5.5 | 86.8% | – | 2026-11-05 |
| 9 | GPT-6 Luna | 86.8% | – | 2026-11-05 |
| 10 |  Qwen3.8 Maxmax | 85.0% | 58.8 | 2026-11-05 |
| 11 | Gemini 3.8 Flash | 80.4% | 54.4 | 2026-11-04 |
| 12 | Claude Haiku 5.5 | 80.3% | – | 2026-11-05 |
| 13 | Gemini 3.5 Flash Lite | 69.0% | 51.6 | 2026-11-04 |
| 14 | Gemma 4 26B A4Bopen ↗ | 66.4% | 56.1 | 2026-11-05 |
| 15 | gpt-oss-120bopen ↗ | 57.6% | 45.5 | 2026-11-05 |
More agents & tools benchmarks · All LLM benchmarks
Related benchmarks
Top head-to-heads on FACTS Search
Cite as: BenchLeader, “FACTS Search leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/kaggle_facts_search.
FACTS Search: questions
- What does FACTS Search measure?
- Using web search as a tool and synthesising what comes back without inventing anything. Published by Google. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on FACTS Search?
- As of 6 Nov 2026, the highest score on FACTS Search is 93.3% by GPT-6 Astra (result dated 5 Nov 2026), ahead of GPT-6.1 Sol at 92.8%. Scores come from Google and are re-read daily.
- Which AI model leads FACTS Search?
- GPT-6 Astra leads FACTS Search with 93.3% as of 6 Nov 2026 (result dated 5 Nov 2026), ahead of GPT-6.1 Sol at 92.8%.
- How many models have FACTS Search results?
- 15 model configurations have a FACTS Search result on BenchLeader, all taken from Google via Kaggle Benchmarks.
- Who runs FACTS Search and how often is it updated?
- FACTS Search is published by Google. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does FACTS Search count toward the BenchLeader Index?
- No. FACTS Search is shown for reference but left out of the composite index.