← Web Search LLM Leaderboard
Claude Haiku 5.5
Anthropic · closed weights · data as of
Claude Haiku 5.5 ranks 8 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 GLM 5.3 Flash.
Score with Firecrawl Alexandria61.9
+44.4 points of lift, from 17.5 without050100
Without With Firecrawl Alexandria
Claude Haiku 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 | 1.5 | 43.0 | +41.5 | #8 | −13.4 |
| DeepSearchQA | 24.3 | 70.3 | +46.0 | #9 | −9.9 |
| HLE | 32.4 | 41.6 | +9.3 | #8 | −6.4 |
| Firecrawl Search Bench | 11.8 | 92.6 | +80.8 | #9 | −1.3 |
Cost and speed
$165 per 1,000 questions, model tokens plus Firecrawl credits: 0.7× the median model, 2nd cheapest of 10.
- Time per question
- 156 s
- average, 4th fastest of 10
- Median time
- 49 s
- set-weighted, half take longer
- 90th percentile time
- 511 s
- set-weighted, 1 in 10 take longer
- Failed
- 0.1%
- of questions, scored 0
Claude Haiku 5.5 vs its closest rivals
- GLM 5.3 Flash scores 61.9, within the margin of error. It costs about the same and takes 755 s per question on average.
- MiniMax M3.1 Flash scores 64.8, clearly ahead. It costs 1.2× as much and takes 407 s per question on average.
- Mercury 3 scores 68.7, clearly ahead. It costs 1.5× as much and takes 214 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