← Web Search LLM Leaderboard
Claude Opus 5.5
Anthropic · closed weights · data as of
Claude Opus 5.5 ranks 1 of 10 on the Web Search LLM Leaderboard, across 4,703 questions from BrowseComp, DeepSearchQA, HLE and Firecrawl Search Bench.
Score with Firecrawl Alexandria79.0
+41.9 points of lift, from 37.1 without050100
Without With Firecrawl Alexandria
Claude Opus 5.5 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 | 26.1 | 73.8 | +47.8 | #4 | +17.4 |
| DeepSearchQA | 48.6 | 87.2 | +38.5 | #1 | +6.9 |
| HLE | 56.4 | 59.4 | +3.1 | #1 | +11.4 |
| Firecrawl Search Bench | 17.3 | 95.4 | +78.1 | #3 | +1.5 |
Cost and speed
$372 per 1,000 questions, model tokens plus Firecrawl credits: 1.7× the median model, 8th cheapest of 10.
- Time per question
- 126 s
- average, 2nd fastest of 10
- Median time
- 28 s
- set-weighted, half take longer
- 90th percentile time
- 355 s
- set-weighted, 1 in 10 take longer
- Failed
- 0.2%
- of questions, scored 0
Claude Opus 5.5 vs its closest rivals
- GPT-6.1 Sol scores 77.9, clearly behind. It is 1.8× cheaper and takes 134 s per question on average.
- Kimi K3.1 Pro scores 77.4, clearly behind. It costs 1.4× as much and takes 386 s per question on average.
- Pareto 26.9 scores 77.1, clearly behind. It costs 1.8× as much and takes 328 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