← Web Search LLM Leaderboard
Pareto 26.9
Unbiased · closed weights · data as of
Pareto 26.9 ranks 2 of 10 on the Web Search LLM Leaderboard, across 4,703 questions from BrowseComp, DeepSearchQA, HLE and Firecrawl Search Bench. It shares that rank with GPT-6.1 Sol and Kimi K3.1 Pro.
Score with Firecrawl Alexandria77.1
+44.7 points of lift, from 32.5 without050100
Without With Firecrawl Alexandria
Pareto 26.9 on each question set
Points out of 100. Rank is among the 10 models on that set, by score with Firecrawl Alexandria; vs median is the gap to the median model there. Firecrawl Search Bench is our own question set; its questions are not published.
| Question set | Without | WithWith Firecrawl Alexandria | Lift | Rank | vs median |
|---|---|---|---|---|---|
| BrowseComp | 27.7 | 78.0 | +50.2 | #1 | +21.5 |
| DeepSearchQA | 36.5 | 84.9 | +48.4 | #3 | +4.6 |
| HLE | 47.7 | 49.9 | +2.2 | #5 | +1.9 |
| Firecrawl Search Bench | 18.1 | 95.9 | +77.8 | #2 | +2.0 |
Cost and speed
$658 per 1,000 questions, model tokens plus Firecrawl credits: 2.9× the median model, most expensive of 10.
- Time per question
- 328 s
- average, 6th fastest of 10
- Median time
- 162 s
- set-weighted, half take longer
- 90th percentile time
- 936 s
- set-weighted, 1 in 10 take longer
- Failed
- 0.2%
- of questions, scored 0
Pareto 26.9 vs its closest rivals
- Kimi K3.1 Pro scores 77.4, within the margin of error. It is 1.2× cheaper and takes 386 s per question on average.
- GPT-6.1 Sol scores 77.9, within the margin of error. It is 3.2× cheaper and takes 134 s per question on average.
- Claude Opus 5.5 scores 79.0, clearly ahead. It is 1.8× cheaper and takes 126 s per question on average.
- Ranks count only the models clearly ahead, so models too close to call share a rank. How we tested →
- What score and lift mean →What is Firecrawl Alexandria? →Firecrawl search docs →
//
Get started
//
Ready to build?
Start getting web context for free and scale seamlessly as your project expands. No credit card needed.
Are you an AI agent? See setup options