# Vals Web Search Index (Web Search Index)

> A Vals AI comparison of native provider search and Exa across finance-analysis and legal-research tasks.

Canonical page: https://benchlm.ai/benchmarks/valswebsearchindex

- Category: [Agentic](/agentic)
- Last updated: July 16, 2026

## About Web Search Index

- Year: 2026
- Tasks: Finance Agent Benchmark v2 and Legal Research Benchmark tasks
- Format: Accuracy by model and search-tool combination
- Difficulty: Professional web research with controlled search-tool variants
- Paper: [Web Search Index](https://www.vals.ai/benchmarks/web_search)

The index holds each model and the rest of the agent harness constant while swapping the search tool. We mirror the native and Exa rows, task scores, uncertainty, latency, and cost per task. The private dataset and tool-specific runs keep the table outside weighted model rankings.

Web Search Index is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (8 models)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Claude Fable 5 Exa](https://www.vals.ai/models/anthropic_claude-fable-5-exa) | Exa · Exa | Anthropic | 48.45% |
| 2 | [Claude Fable 5](/models/claude-fable) | Native · Native | Anthropic | 46.94% |
| 3 | [GPT-5.6 Sol Exa](https://www.vals.ai/models/openai_gpt-5.6-sol-exa) | Exa · max reasoning · Exa | OpenAI | 45.24% |
| 4 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | Native · max reasoning · Native | OpenAI | 43.58% |
| 5 | [Gemini 3.5 Flash Exa](https://www.vals.ai/models/google_gemini-3.5-flash-exa) | Exa · high reasoning · Exa | Google | 41.36% |
| 6 | [Grok 4.5 Exa](https://www.vals.ai/models/grok_grok-4.5-exa) | Exa · high reasoning · Exa | xAI | 38.75% |
| 7 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | Native · high reasoning · Native | Google | 37.04% |
| 8 | [Grok 4.5](/models/grok-4-5) | Native · high reasoning · Native | xAI | 35.54% |

## FAQ

### What does Web Search Index measure?

A Vals AI comparison of native provider search and Exa across finance-analysis and legal-research tasks.

### Which model leads the published Web Search Index snapshot?

Claude Fable 5 Exa currently leads the published Web Search Index snapshot with a score of 48.45%.

### How many models are evaluated on Web Search Index?

The July 16, 2026 contains 8 AI models.

### Does Web Search Index affect BenchLM's overall score?

Not directly. Web Search Index is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.