Best web search API for RAG by developer retrieval accuracy: Perplexity, 77.3% task completion in this benchmark. Fastest search: Parallel turbo, 333ms (64.7% task completion). Lowest recorded search price: TinyFish, $0.00 per 1,000 queries within its free-tier limits.
Compare web search APIs for retrieval-augmented generation using our developer retrieval benchmark, with accuracy, search speed and pricing per 1,000 queries. Compare 10 providers across 14 configurations, including Exa, Tavily, Brave, Parallel and Firecrawl, on the same questions.
Updated 20 September 2026. Last measured 15 September 2026.
Task completion on 100 developer questions with the same agent using search results. Each task requires a passing patch and a citation from the run's sources. Search latency and total task latency are shown separately. Prices are the benchmark-recorded rate basis, with the same configurations used for the quality results.
| Rank | Provider & configuration | Task completion | Mean search time | Recorded $ / 1,000 queries | Median task time | Median task cost | Official docs |
|---|---|---|---|---|---|---|---|
| 1 | Perplexitysearch_context_size=low/search | 77.3% | 957ms | $5.00Search API · POST /search | 18.31s | $0.06 | Official docs ↗ |
| 2 | Firecrawlfirecrawl-search/v2/search | 70.3% | 2.87s | $5.002 credits / 10 results | 27.95s | $0.06 | Official docs ↗ |
| 3 | Parallel fastmode=fast/v1/search | 66.7% | 953ms | $1.00mode=fast · 10 results | 22.09s | $0.05 | Official docs ↗ |
| 4 | Exa fasttype=fast/search | 66.3% | 626ms | $7.00type=fast · up to 10 results | 19.97s | $0.10 | Official docs ↗ |
| 5 | Parallel turbomode=turbo/v1/search | 64.7% | 333ms | $1.00mode=turbo · 10 results | 18.69s | $0.06 | Official docs ↗ |
| 6 | Exa instanttype=instant/search | 61.3% | 447ms | $7.00type=instant · up to 10 results | 20.58s | $0.10 | Official docs ↗ |
| 7 | TinyFishfree · 30 req/min capapi.search.tinyfish.ai | 59.3% | 2.15s | $0.0030 req/min cap | 27.84s | $0.03 | Official docs ↗ |
| 8 | Tavily basicsearch_depth=basic/search | 51.0% | 1.72s | $8.001 credit · $0.008 PAYG | 26.40s | $0.10 | Official docs ↗ |
| 9 | Nimblesearch_depth=lite · full_content=false · focus=general/v2/search | 46.7% | 3.15s | $1.10search_depth=lite · full_content=false | 32.72s | $0.04 | Official docs ↗ |
| 10 | Linkup fastdepth=fast · outputType=searchResults/v1/search | 43.3% | 1.39s | $5.00depth=fast · searchResults | 26.76s | $0.10 | Official docs ↗ |
| 11 | Nimblesearch_depth=standard · full_content=false · focus=general/v2/search | 42.3% | 878ms | $5.00search_depth=standard · full_content=false | 25.27s | $0.08 | Official docs ↗ |
| 12 | Youextraction_mode=highlights/v1/search | 41.3% | 596ms | $5.00highlights included in Web Search | 23.93s | $0.10 | Official docs ↗ |
| 13 | Youextraction_mode=highlights · knowledge=core/v1/search | 38.3% | 677ms | $5.00highlights · knowledge=core · same Web Search price assumption | 21.05s | $0.10 | Official docs ↗ |
| 14 | Bravecount=10 · result_filter=web/res/v1/web/search | 38.0% | 523ms | $5.00Recorded search-request rate; see official terms. | 19.87s | $0.08 | Official docs ↗ |
| Stage | Keep or measure | How to use this comparison |
|---|---|---|
| Retrieve | Query, endpoint mode, result URLs, passages and search latency | Compare the same search configurations and request times in the main table. |
| Select context | Evidence coverage, source diversity and tokens passed to the model | Test your chunking and reranking on the same retrieved results. |
| Generate and cite | Answer correctness and support for each cited claim | Developer task completion measures successful evidence use in the tested agent. Evaluate your own answer format separately. |
| Operate | Search calls per answer, model spend and end-to-end time | Keep per-query search pricing separate from the full application bill. |
For direct factual retrieval, also see the search accuracy comparison. For an agent that assembles evidence across many searches, see the multi-search comparison.
The main table measures a developer agent that retrieves documentation and uses that evidence to solve a task. Accuracy here means grounded task completion: a patch must pass its checks and cite a source from the run. That is useful evidence for technical RAG and agentic retrieval. General chatbot answer quality still needs evaluation on your own questions and generation pipeline.
Retain the source URL, title, passage and retrieval time together. Feed evidence into the answer prompt with stable citation identifiers, then check whether each final claim is supported by its cited passage. A high search score is one part of this pipeline; chunk selection, context limits and the answer model also influence the result.
Mean search time measures the API call. Median task time includes repeated searches and the agent’s work. For a RAG assistant, measure retrieval, reranking and generation separately, then track end-to-end time. Choose a configuration that meets the accuracy level your application needs before trading quality for a shorter response.
Search cost per answer equals the number of search calls multiplied by the per-call price. Add input and output tokens, reranking and any scraping. The table separates recorded search price per 1,000 queries from median full-task cost, so a cheap API request and a cheap completed answer remain distinct.
Build a held-out set covering fresh facts, ambiguous questions, missing evidence and conflicting sources. Inspect retrieved passages, citation support and final answer correctness separately. Keep the model and prompt fixed while swapping search providers. Then test the chosen provider at your expected concurrency and context budget.
Task completion on 100 developer questions with the same agent using search results. Each task requires a passing patch and a citation from the run's sources. Search latency and total task latency are shown separately. Each row is one measured configuration. Missing measurements appear as a dash.
Open benchmark code ↗ · Public dataset sample ↗ · All benchmark methods →
Compare the developer search and search-plus-scrape boards →
Best web search API for RAG by developer retrieval accuracy: Perplexity, 77.3% task completion in this benchmark. Fastest search: Parallel turbo, 333ms (64.7% task completion). Lowest recorded search price: TinyFish, $0.00 per 1,000 queries within its free-tier limits.
Task completion on 100 developer questions with the same agent using search results. Each task requires a passing patch and a citation from the run's sources. Search latency and total task latency are shown separately.
The recorded search endpoint's request price. Model-token spend, optional scraping and plan conditions are separate. For rows with task measurements, median task cost is shown in its own column. Check the linked provider terms for current pricing.
The main ranking measures grounded developer task completion with a fixed agent and search results. Use it to shortlist retrieval providers, then measure final answer correctness and citation support on your own RAG questions.