The most accurate web search API in the world
SpecStream leads WideSearch and SimpleQA Verified, the standard benchmarks for search agents. On WideSearch, even a low-cost model on SpecStream beats every other search API running on a frontier model, at a fraction of the cost.
Why SpecStream leads
No other search API works the way ours does. SpecStream reads the live web at the moment you ask, goes to each site by its own road, and hands the agent the evidence with its source.
WideSearch
WideSearch, from ByteDance Seed, gives an AI agent broad research tasks and scores how much of the answer it collects correctly.
| Provider | Agent | Score | Cost per 1,000 tasks |
|---|---|---|---|
| SpecStream | GPT-5.6 Luna | 60.9 | $39.40 |
| Parallel Advanced | GPT-5.6 Sol | 57.6 | $692 |
| Exa Auto | GPT-5.6 Sol | 55.9 | $1,061 |
| Tavily | GPT-5.6 Sol | 55.9 | $1,072 |
| Parallel Basic | GPT-5.6 Sol | 55.3 | $965 |
| Perplexity | GPT-5.6 Sol | 53.5 | $547 |
| Exa Auto | GPT-5.6 Luna | 53.0 | $41.20 |
| Tavily | GPT-5.6 Luna | 47.9 | $107 |
| Perplexity | GPT-5.6 Luna | 47.0 | $24.80 |
| Parallel Fast | GPT-5.6 Luna | 45.5 | $10.50 |
| Parallel Turbo | GPT-5.6 Luna | 44.0 | $10.10 |
SimpleQA Verified
SimpleQA Verified asks short factual questions with one verified answer each.
| Provider | Agent | Accuracy |
|---|---|---|
| SpecStream | GPT-5.6 Sol | 99 |
| Parallel Advanced | GPT-5.6 Sol | 97 |
| Parallel Basic | GPT-5.6 Sol | 97 |
| SpecStream | GPT-5.6 Luna | 96 |
| Perplexity | GPT-5.6 Sol | 95 |
| Parallel Fast | GPT-5.6 Luna | 94 |
| Perplexity | GPT-5.6 Luna | 94 |
| Tavily | GPT-5.6 Luna | 94 |
| Tavily | GPT-5.6 Sol | 92 |
| Parallel Turbo | GPT-5.6 Luna | 91 |
| Exa Auto | GPT-5.6 Luna | 91 |
| Exa Auto | GPT-5.6 Sol | 91 |
Method
- WideSearch by ByteDance Seed. Item F1 on a 100-task sample, 50 in English and 50 in Chinese, on GPT-5.6 Luna at low reasoning, with the benchmark's own scorer. Cost covers model tokens and tool calls at API rates. Run November 5, 2026.
- SimpleQA Verified. Accuracy on 100 questions, on GPT-5.6 Luna at low reasoning and GPT-5.6 Sol at high reasoning. Run November 5, 2026.
Run it on your own work
Sign up for $10 of work, no card, and run your first request in the workspace, through the API or over MCP.