Web Search Index leaderboard
As of 6 Nov 2026, the top score on Web Search Index is 49.1% by GPT-5.6 Sol Keenable (OpenAI), ahead of Claude Fable 5 Exa at 48.5%, across 9 model configurations with a published result (Vals AI, result dated 21 Oct 2026). Questions whose answers require searching the web; models run with their native search and with an independent search tool. Web Search Index is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- GPT-5.6 Sol Keenable holds the top Web Search Index score at 49.1% (Vals AI, 21 Oct 2026).
- Claude Fable 5 Exa is second on Web Search Index at 48.5% (Vals AI, 21 Oct 2026).
- Web Search Index: Answering questions that need live web search, with native and independent tools. Run by Vals AI (Vals AI).
- 9 model configurations have a published Web Search Index result on BenchLeader (BenchLeader, 6 Nov 2026).
- Web Search Index does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Answering questions that need live web search, with native and independent tools. Run by Vals AI.
- Published by
- Vals AI
- Category
- Agents & tools
- Index weight
- Reference only
- Models
- 9
- Data as of
- 6 Nov 2026
Vals AI (vals.ai).
What the test looks like
Questions whose answers require searching the web; models run with their native search and with an independent search tool.
How it is scored
Accuracy, run by Vals AI.
What to keep in mind
Results depend on the search backend and on the day the question was asked.
- 1GPT-5.6 Sol Keenable (max)49.1%
- 2Claude Fable 5 Exa48.5%
- 3Claude Fable 546.9%
- 4GPT 5.6 Sol Exa (max)45.2%
- 5GPT-5.6 Sol (max)43.6%
- 6Gemini 3.5 Flash Exa (high)41.4%
- 7Grok 4.5 Exa (high)38.8%
- 8Gemini 3.5 Flash (high)37.0%
- 9Grok 4.5 (high)35.5%
| # | |||||
|---|---|---|---|---|---|
| 1 |  GPT-5.6 Sol Keenablemax | 49.1% | – | $8.84 | 2026-10-21 |
| 2 |  Claude Fable 5 Exa | 48.5% | – | $3.48 | 2026-10-21 |
| 3 | Claude Fable 5 | 46.9% | 66.5 | $4.37 | 2026-10-21 |
| 4 | GPT 5.6 Sol Examax | 45.2% | – | $18.97 | 2026-10-21 |
| 5 | GPT-5.6 Solmax | 43.6% | 65.4 | $0.954 | 2026-10-21 |
| 6 |  Gemini 3.5 Flash Exahigh | 41.4% | – | $1.18 | 2026-10-21 |
| 7 |  Grok 4.5 Exahigh | 38.8% | – | $0.576 | 2026-10-21 |
| 8 | Gemini 3.5 Flashhigh | 37.0% | 62.5 | $0.499 | 2026-10-21 |
| 9 | Grok 4.5high | 35.5% | 56.9 | $0.991 | 2026-10-21 |
More agents & tools benchmarks · All LLM benchmarks
Related benchmarks
Top head-to-heads on Web Search Index
Cite as: BenchLeader, “Web Search Index leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/vals_web_search.
Web Search Index: questions
- What does Web Search Index measure?
- Questions whose answers require searching the web; models run with their native search and with an independent search tool. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on Web Search Index?
- As of 6 Nov 2026, the highest score on Web Search Index is 49.1% by GPT-5.6 Sol Keenable (result dated 21 Oct 2026), ahead of Claude Fable 5 Exa at 48.5%. Scores come from Vals AI and are re-read daily.
- Which AI model leads Web Search Index?
- GPT-5.6 Sol Keenable leads Web Search Index with 49.1% as of 6 Nov 2026 (result dated 21 Oct 2026), ahead of Claude Fable 5 Exa at 48.5%.
- How many models have Web Search Index results?
- 9 model configurations have a Web Search Index result on BenchLeader, all taken from Vals AI.
- Who runs Web Search Index and how often is it updated?
- Web Search Index is published by Vals AI. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does Web Search Index count toward the BenchLeader Index?
- No. Web Search Index is shown for reference but left out of the composite index.