Best software · Developer & data
10 Best Search APIs for AI of 2026
This list is for developers who give AI agents, chatbots and RAG apps live web search results. It covers AI-native search indexes (Exa, Brave Search API, Parallel, Perplexity, You.com, Linkup), agent web toolkits (Tavily, Firecrawl, Jina Reader) and a search engine results API (SerpApi). We compared documented features and published prices for 10 APIs. Scores weigh features at 40%, ease of use at 30% and value at 30%.
By the Mojolin research desk · Updated November 12, 2026 · 10 tools compared · Prices checked on vendor pages, latest November 12, 2026
Questions or corrections: editorial@mojolin.com
Mojolin is self-funded How we score
Exa is our top pick: Lite search costs $4 per 1,000 requests, the free Starter plan gives $10 of credits a month, and the same key covers page contents, monitors and specialized people, company and scholarly indexes. Tavily is the easiest to adopt, with 1,000 free credits a month and one credit per basic search. Brave Search API serves its own independent index at 50 queries per second for $5 per 1,000 requests, and Parallel and Perplexity Fast Search are the lowest-cost options at $1 per 1,000 requests. Pick SerpApi when you need Google results specifically, and Firecrawl or Jina Reader when reading full pages matters as much as search.
Top picks
Best overall
Exa
9.0 / 10
Developers building agents that need fast, configurable web search with page text
Read the review →Runner-up
Brave Search API
8.9 / 10
Teams that want results from an independent index at high query rates
Read the review →Also great
Parallel Search API
8.9 / 10
Cost-sensitive teams running high volumes of agent search calls
Read the review →Comparison table
| # | Tool | Pricing | Score | |
|---|---|---|---|---|
| 1 | Exa | Developers building agents that need fast, configurable web search with page text | Free Starter plan with $10 of credits per month; Search from $4 per 1,000 requests pay as you go; Contents $1 per 1,000 pages | 9.0 |
| 2 | Brave Search API | Teams that want results from an independent index at high query rates | Prepaid credits with $5 of free usage per month; Search plan $5 per 1,000 requests; Answers $4 per 1,000 requests plus $5 per million tokens | 8.9 |
| 3 | Parallel Search API | Cost-sensitive teams running high volumes of agent search calls | Pay as you go with $5 free each month and up to $80 at signup; Search from $1 per 1,000 requests; Scale plan $1,000 per month in credits | 8.9 |
| 4 |  Tavily | Agent developers who want a free monthly allowance and simple credit pricing | Free Researcher plan with 1,000 credits per month; pay as you go at $0.008 per credit; Project plan $30 per month for 4,000 credits | 8.9 |
| 5 |  Perplexity Search API | Developers who want raw ranked results with page text at a flat per-request price | Pay as you go; Search API $5 per 1,000 requests, or $1 per 1,000 with Fast Search; no token charges | 8.8 |
| 6 |  Firecrawl | Developers who need search results and full page scraping in one product | Free plan with 1,000 credits per month; Hobby $19 per month billed monthly or $16 billed annually for 5,000 credits; Standard $99 per month | 8.6 |
| 7 |  You.com API | Developers who want search, cited answers and research endpoints from one vendor | Pay per call; Web Search API $5 per 1,000 calls; Contents API $1 per 1,000 pages; Answer API $5 per 1,000 calls | 8.6 |
| 8 |  Linkup | Teams that want flat per-request pricing with SOC 2 Type II on every plan | Pay as you go; 4,000 free queries to start; Search $0.005 to $0.006 per request; Fetch $0.001 to $0.006; Research $0.25 to $2.50 | 8.5 |
| 9 |  Jina Reader | Developers who want a free, URL-prefix way to read pages and run simple searches | Free to start with token-based paid keys; See vendor pricing (USD check pending) | 8.3 |
| 10 |  SerpApi | Teams that need Google results, AI Overviews or vertical engines as structured data | Free plan with 250 searches per month; Starter $25 per month for 1,000 searches; Developer $75; Production $150 | 8.3 |
Scoring weights
Every tool gets three editorial scores out of 10, from vendor documentation, product pages and the published price. The overall score is the weighted average below, the same method on every list. Full method
The full ranking
Rank 1 · Best overall
Exa
A web search API built for AI agents, with search, page contents, monitors and an asynchronous research agent.
exa.ai9.0
out of 10
Best for: Developers building agents that need fast, configurable web search with page text
Exa is a search engine designed to be called by software rather than people. Its Search endpoint comes in three tiers, Lite at $4, Auto at $7 and Deep at $12 per 1,000 requests, each returning up to 10 results with page text and highlights, and Exa documents latency from 450ms to 1s. Beyond the main web index it runs people, company and scholarly works indexes, which matters if your agent looks up prospects or papers rather than general pages.
The rest of the product fills the gaps around search: Contents fetches full pages at $1 per 1,000, Monitors re-run a search on a schedule and send webhooks, and the Agent API runs asynchronous research with effort levels from $0.012 to $1.00 per run. The free Starter plan includes $10 of credits each month plus a $10 onboarding bonus and needs no card. The catch is that costs climb with Deep search and extra results, and zero data retention is only on Enterprise.
Standout feature
Exa is the only API here that pairs tiered web search with dedicated people, company and scholarly works indexes.
Use cases
Coding agent builders
Look up current docs, repos and changelogs during a code generation loop
Outcome: Answers grounded in current package and API references
Sales tooling teams
Query the company and people indexes to enrich leads
Outcome: Prospect records filled from live web data
News products
Run Monitors on a daily schedule with webhooks
Outcome: Fresh events delivered without polling
Key features
- Lite, Auto and Deep search at $4, $7 and $12 per 1,000 requests with up to 10 results
- Contents endpoint returns full page text and highlights at $1 per 1,000 pages
- Separate web, people, company and scholarly works indexes
- Agent API for asynchronous research runs from $0.012 to $1.00 per run
- MCP server, Claude Connector and 50+ integrations
Pros
- Starter plan gives $10 of credits every month with no payment method
- Configurable latency from 450ms to 1s on search
- Pay-as-you-go Developer plan raises the limit to 25 search queries per second
Cons
- Deep search costs three times the Lite rate
- Zero data retention needs an Enterprise plan with a $12,000 annual commitment
- Each result above 10 adds $1 per 1,000 requests
Rank 2 · Runner-up
Brave Search API
Web search results from Brave's own independent index, with an LLM Context endpoint and an Answers API.
brave.com8.9
out of 10
Best for: Teams that want results from an independent index at high query rates
Brave Search API stands apart because every result comes from Brave's own index, crawled and ranked by Brave, with no fallback to another engine. For developers that means one key returns web, news, image, video and places results, plus an LLM Context endpoint that hands your model ranked, extracted page text sized to a token budget. Goggles let you rerank and filter results with your own rules.
Pricing is prepaid credits rather than a subscription. The Search plan costs $5 per 1,000 requests and allows 50 queries per second, and the Answers plan, which returns cited answers and works with the OpenAI SDK, costs $4 per 1,000 requests plus $5 per million tokens at 2 queries per second. Each plan includes $5 of free credits a month, though Brave asks for a credit card to prevent abuse. Rights to store results or train models require a custom plan, which matters for dataset builders.
Standout feature
Brave Search API is the only option here that serves results entirely from its own independent web index.
Use cases
Chatbot teams
Ground answers with the LLM Context endpoint
Outcome: Compact, ranked page text sized for the model
High-traffic assistants
Serve live search at up to 50 queries per second
Outcome: No queueing at peak load
Local and travel apps
Pull places and news results from the same key
Outcome: One integration for several result types
Key features
- Results from Brave's own crawler and index with no Google or Bing fallback
- LLM Context endpoint that returns ranked, extracted page text for a model
- Web, news, images, videos and places results from one API key
- Goggles for custom reranking and result filtering
- Answers plan with cited, streamed answers and OpenAI SDK compatibility
Pros
- Search plan allows 50 queries per second
- $5 of free credits every month on each plan
- Prepaid credits with no subscription
Cons
- A credit card is required even for free usage
- Answers plan is capped at 2 queries per second
- Storing results or using them to train models needs a custom plan
Rank 3 · Also great
Parallel Search API
Pay-as-you-go web APIs for agents, covering Search, Extract, Task, Monitor and more.
parallel.ai8.9
out of 10
Best for: Cost-sensitive teams running high volumes of agent search calls
Parallel sells a family of web APIs priced per unit of work rather than per token. Search starts at $1 per 1,000 requests with 10 results included, the lowest published entry price among the general search APIs we compared, and extra results cost $1 per 1,000. Around it sit Extract at $1 per 1,000 URLs, a Task API for research from $5 per 1,000 successful runs, and Monitor from $3 per 1,000 checks, so an agent can search, read and track pages from one account.
The Build plan has no subscription: you add a card, get up to $80 at signup and $5 of free usage every month, and pay published rates after that. Larger teams can buy the Scale plan at $1,000 a month in credits, or $10,000 a year for $12,000 in credits. The trade-off is complexity, since each API has its own tiers, so budget a little time to model costs before you commit.
Standout feature
Parallel publishes the lowest general web search starting price on this list at $1 per 1,000 requests.
Use cases
Agent platforms
Run millions of search tool calls a month
Outcome: Lower search cost per agent step
Data enrichment teams
Use the Task API to research and fill company records
Outcome: Structured fields with web sources
Monitoring products
Watch pages and companies with the Monitor API
Outcome: Alerts when something changes
Key features
- Search API from $1 per 1,000 requests with 10 results included
- Extract API at $1 per 1,000 URLs
- Task API for research runs from $5 per 1,000 successful runs
- Monitor API from $3 per 1,000 checks
- Specialized data connectors and 100+ integrations
Pros
- Lowest published search starting price on this list at $1 per 1,000 requests
- Recurring $5 monthly free allowance plus up to $80 at signup
- SOC 2 Type II covers Build, Scale and Enterprise plans
Cons
- Several APIs and price tiers take time to compare
- Scale plan credits do not roll over month to month
- Startup grants are smaller than some rivals at up to $250
Rank 4
Tavily
A web access API for AI agents with Search, Extract, Crawl, Map and Research endpoints billed in credits.
tavily.com8.9
out of 10
Best for: Agent developers who want a free monthly allowance and simple credit pricing
Tavily packages web access for agents into five endpoints: Search for results, Extract for clean page text, Crawl for whole sites, Map for URL discovery and Research for multi-step reports. Everything is billed in API credits, and the docs make the math easy to follow: a basic search costs 1 credit, five successful basic extractions cost 1 credit, and advanced modes cost more.
The free Researcher plan includes 1,000 credits a month with no credit card, which covers a prototype, and credits reset on the first day of each month. After that you can pay as you go at $0.008 per credit or pick the Project plan at $30 a month for 4,000 credits and higher rate limits. That makes a basic search about $8 per 1,000, more than Exa Lite or Brave, but the single key for search, extraction and crawling keeps integration short. Tavily says more than 3 million developers use it, and it offers 100,000 free credits to startups in its program.
Standout feature
Tavily's free plan of 1,000 credits a month covers about 1,000 basic searches with no credit card.
Use cases
Hackathon and prototype teams
Add live search to an agent on the free 1,000 credit plan
Outcome: A working demo with no card on file
RAG developers
Search, then Extract the top pages into clean text
Outcome: Grounded answers with source content
Documentation tools
Crawl and Map a vendor site
Outcome: A full URL list and page text for indexing
Key features
- Search endpoint where a basic search costs 1 API credit
- Extract returns clean page content, with 5 basic extractions per credit
- Crawl and Map endpoints for whole sites and URL discovery
- Research endpoint for in-depth multi-step web research
- Python and JavaScript SDKs
Pros
- 1,000 free credits every month with no credit card
- Pay as you go at $0.008 per credit with no subscription
- Startup program offers 100,000 free credits
Cons
- Free and plan credits stop working once used up until the monthly reset
- Advanced search and extraction use more credits per call
- Enterprise security terms need a sales conversation
Rank 5
Perplexity Search API
Ranked web results as JSON from Perplexity, with extracted page text and no token billing.
perplexity.ai8.8
out of 10
Best for: Developers who want raw ranked results with page text at a flat per-request price
Perplexity's Search API returns ranked web results as JSON with page text already extracted into each result's snippet. There is no model and no generated answer, so billing is simple: $5 per 1,000 requests for up to 20 results, or $1 per 1,000 with Fast Search, with no token charges. Changing how much text you extract does not change the price, and a query array of five queries in one POST counts as one request, which keeps batch lookups cheap.
You can filter by domain, language and date and bias results toward a country. Perplexity does not bill invalid, rate-limited or failed requests, though a successful request with no results is still charged. If you want a written, cited answer instead of results, Perplexity steers developers to its Agent API, which adds model token costs and per-tool fees such as $0.0005 per fetch_url call.
Standout feature
Perplexity Search API bills a query array of five queries as one request, with no token charges.
Use cases
RAG pipelines
Fetch ranked results with extracted text for your own model
Outcome: Context without a separate scraping step
Batch research jobs
Send five queries in one request array
Outcome: Lower cost per lookup
Regional apps
Filter by language and bias results by country
Outcome: Results that match the user's market
Key features
- Ranked web results as JSON with page text extracted into each snippet
- Up to 20 results per request at the standard price
- Filters by domain, language and date, with country biasing
- Fast Search mode at $1 per 1,000 requests
- Agent API with web search, fetch_url and other tools for full answers
Pros
- Flat per-request price with no token billing
- Invalid, rate-limited and failed requests are not billed
- Amount of extracted text does not change the price
Cons
- Successful requests are billed even when they return no results
- No permanent free plan is listed on the pricing page
- Generated answers need the separate Agent API with token costs
Rank 6
Firecrawl
A web data API for AI that combines search with scraping, crawling and mapping on one credit balance.
firecrawl.dev8.6
out of 10
Best for: Developers who need search results and full page scraping in one product
Firecrawl started as a scraper for AI apps and now includes a search endpoint, so it suits agents that search and then read pages in depth. Credits are shared across endpoints: search costs 2 credits per 10 results, and scrape, crawl and map cost 1 credit per page or call, with structured JSON output adding 4 credits per page. The free plan gives 1,000 credits a month, enough for 500 searches or 1,000 pages.
Paid plans are monthly subscriptions. Hobby costs $19 a month, or $16 billed annually, for 5,000 credits; Standard is $99 a month for 100,000 credits with 500 search requests a minute. Paid plans can top up in $5 increments when credits run out. For search alone, per-request APIs like Parallel or Perplexity cost less, but Firecrawl saves you from wiring a second vendor once your agent needs full page content.
Standout feature
Firecrawl lets search results and scraped pages draw on one shared credit balance.
Use cases
AI app developers
Search, then scrape the top results to markdown
Outcome: Full page context from one API
Knowledge base builders
Crawl and Map documentation sites
Outcome: Complete page sets for indexing
Side projects
Prototype on the free 1,000 credits
Outcome: No cost until usage grows
Key features
- Search costs 2 credits per 10 results
- Scrape, Crawl and Map at 1 credit per page or call
- Interact for browser sessions at 2 credits per browser minute
- Search results and scraped pages draw on one credit balance
- Pay-as-you-go top ups in $5 increments on paid plans
Pros
- Free plan includes 1,000 credits a month with no card
- One balance covers both search and scraping
- Standard plan allows 500 search requests a minute
Cons
- Monthly plans are the main way to buy, unlike pure pay-as-you-go rivals
- JSON and question formats add 4 credits per page
- Free plan is limited to 2 concurrent requests
Rank 7
You.com API
Web Search, Contents, Answer and Research APIs from You.com, plus an inference router grounded in search.
you.com8.6
out of 10
Best for: Developers who want search, cited answers and research endpoints from one vendor
You.com's developer platform covers most of what an agent needs from the web. The Web Search API returns ranked web and news results as JSON at $5 per 1,000 calls, with up to 100 results per call, and can return snippets, highlights or full pages; full page extraction adds $1 per 1,000 pages fetched live, and cached pages are free. The Contents API reads known URLs at the same $1 per 1,000 pages, and the Answer API returns a cited Markdown answer for $5 per 1,000 calls.
For longer tasks there are Research and Finance Research APIs, and a newer Inference Router serves open-weight models grounded in search. The MCP server has a keyless free profile allowing 100 queries a day, which is handy for trying search inside Claude Code or Cursor. The downside is the number of endpoints and add-ons, which makes it worth reading the billing page before you scale.
Standout feature
You.com's Web Search API returns up to 100 results per call for $5 per 1,000 calls.
Use cases
Assistant builders
Call the Answer API for cited responses
Outcome: Answers with verbatim source excerpts
Coding agents
Add the free MCP profile to Claude Code or Cursor
Outcome: Live search without an API key
Finance teams
Run the Finance Research API on filings and earnings
Outcome: Cited reports on numerical questions
Key features
- Web Search API with up to 100 results per call
- Full page extraction add-on at $1 per 1,000 pages fetched live
- Answer API returns a cited Markdown answer at $5 per 1,000 calls
- Research and Finance Research APIs for multi-step cited reports
- MCP server with a keyless free profile for 100 queries a day
Pros
- Up to 100 results per Web Search call
- Pages served from cache are free on full page extraction
- Keyless MCP access lets you test search before signing up
Cons
- No permanent free API plan beyond the MCP profile
- Full page extraction adds up to $0.01 per call at the default 10 results
- Many endpoints and add-ons make costs harder to estimate
Rank 8
Linkup
A web search API for AI with Fast, Standard and Deep search modes, URL fetch and asynchronous research.
linkup.so8.5
out of 10
Best for: Teams that want flat per-request pricing with SOC 2 Type II on every plan
Linkup keeps its offer short: three products, each with a published price per request. Search, in Fast, Standard and Deep modes, costs $0.005 to $0.006 per request with 1 to 3 second latency, Fetch extracts URL content in under 2 seconds for $0.001 to $0.006, and Research runs asynchronous deep research in 1 to 10 minutes for $0.25 to $2.50. New accounts can run 4,000 queries for free.
Compliance is a selling point for enterprise buyers, since SOC 2 Type II is included on all plans, while zero data retention, personalized indexes and dedicated refresh rates sit in Enterprise. Startups can apply for $5,000 in free credits. Linkup has fewer endpoints than Exa or You.com, and per search it costs more than the $1 entry tiers from Parallel and Perplexity, but the flat price band makes budgets easy to set.
Standout feature
Linkup includes SOC 2 Type II on all plans and offers startups up to $5,000 in free credits.
Use cases
Enterprise AI teams
Add web search to an internal assistant
Outcome: SOC 2 Type II coverage from day one
Agent developers
Switch between Fast and Deep modes per task
Outcome: Speed or depth chosen per call
Startups
Apply for $5,000 in free credits
Outcome: Months of search before paying
Key features
- Fast, Standard and Deep search modes for agent tool calls
- Fetch endpoint for URL content extraction in under 2 seconds
- Research endpoint for asynchronous deep research in 1 to 10 minutes
- SOC 2 Type II included on all plans
- Enterprise options for personalized indexes and private environments
Pros
- 4,000 free queries to start
- Narrow search price band of $0.005 to $0.006 per request
- Startups can apply for $5,000 in free credits
Cons
- Search at about $5 to $6 per 1,000 requests costs more than Parallel or Perplexity Fast Search
- Zero data retention requires the Enterprise plan
- No recurring monthly free allowance is listed
Rank 9
Jina Reader
Turns any URL into LLM-friendly text with r.jina.ai and searches the web with s.jina.ai.
jina.ai8.3
out of 10
Best for: Developers who want a free, URL-prefix way to read pages and run simple searches
Jina Reader is the lightest option on this list. Prefix any URL with r.jina.ai and you get clean text an LLM can read; prefix a query with s.jina.ai and it searches the web and returns the top five results with their contents, so a model gets text rather than bare links. Request headers control the browser engine, a token budget, image and link handling, and jina-ocr-v1 parses documents and images.
The Reader endpoint works without an API key at lower rate limits, while search needs a key and allows 100 requests a minute on free and paid keys and 1,000 on premium. Billing is in prepaid tokens, and each search request costs a fixed number of tokens starting at 10,000. Jina notes that you may be charged in USD, EUR or other currencies depending on location, so we could not confirm a USD price from here. Five results per search makes it a supplement rather than a main search API.
Standout feature
Jina Reader turns any URL into LLM-ready text with nothing more than an r.jina.ai prefix.
Use cases
Solo developers
Prefix URLs with r.jina.ai inside a prompt pipeline
Outcome: Clean page text with no scraping code
Agent builders
Use s.jina.ai for quick searches with page contents
Outcome: Five results ready for the model
Document apps
Parse PDFs and images with jina-ocr-v1
Outcome: Text from scanned files
Key features
- r.jina.ai converts a URL into clean, LLM-friendly text
- s.jina.ai searches the web and returns the top five results with their contents
- Options for browser engine, token budget, link and image handling
- jina-ocr-v1 for document and image parsing
- MCP server at mcp.jina.ai
Pros
- Reading a page needs only a URL prefix
- Search returns page contents, not just links
- Reader can be used without an API key at lower rate limits
Cons
- Search requires an API key and each request costs a fixed number of tokens, from 10,000
- Search returns five results per query
- Prices depend on location and may be charged in EUR or other currencies
Rank 10
SerpApi
A search engine results API that returns structured JSON from Google, Bing, DuckDuckGo, Amazon and many more engines.
serpapi.com8.3
out of 10
Best for: Teams that need Google results, AI Overviews or vertical engines as structured data
SerpApi is different from the AI-native indexes on this list: it returns what a search engine shows, parsed into JSON. That makes it the pick when your app needs Google specifically, including Google AI Mode and AI Overview results, or vertical engines such as Maps, Scholar, Shopping, Jobs and Flights, as well as Bing, DuckDuckGo, Baidu and Amazon. SerpApi also offers U.S. Legal Shield, up to $2 million of coverage for scraping and parsing search engine data.
The cost per query is the highest here. The free plan includes 250 searches a month, Starter costs $25 a month for 1,000 searches with 200 an hour throughput, and Developer and Production cost $75 and $150. Results are snippets and links rather than page text, so an agent usually pairs SerpApi with a reader such as Firecrawl or Jina Reader.
Standout feature
SerpApi is the only API here that returns Google AI Overview and AI Mode results as structured JSON.
Use cases
SEO and AI visibility tools
Track Google AI Overview and AI Mode results
Outcome: Structured data on what Google shows
Shopping assistants
Query Google Shopping and Amazon results
Outcome: Live prices and listings
Research apps
Pull Google Scholar results by query
Outcome: Paper titles, links and citations as JSON
Key features
- Google Search, Google AI Mode and Google AI Overview APIs
- Bing, DuckDuckGo, Baidu and Brave search APIs
- Vertical APIs for Google Maps, Scholar, Shopping, Jobs and Flights
- U.S. Legal Shield coverage up to $2 million for scraping and parsing
- ZeroTrace Mode for higher privacy plans
Pros
- Free plan includes 250 searches a month
- Covers Google results that most AI-native indexes do not return
- Month-to-month plans that cancel anytime
Cons
- Starter works out to $25 per 1,000 searches, far above AI-native APIs
- Throughput is capped per hour, 200 an hour on Starter
- Returns search result pages, not page text, so you still need a reader for content
Frequently asked questions
+What is the best search API for AI agents?
Exa is our top pick for most AI agents. Search starts at $4 per 1,000 requests with page text and highlights, and the free Starter plan includes $10 of credits a month. Tavily is the simplest alternative, with 1,000 free credits a month.
+What is the cheapest web search API for LLMs?
Parallel Search API and Perplexity Search API with Fast Search are the cheapest, both from $1 per 1,000 requests on their pricing pages as of November 2026. Parallel includes 10 results per request and a $5 monthly free allowance.
+Is there a free search API for AI?
Yes. Tavily gives 1,000 free credits a month with no credit card, Exa gives $10 of credits a month, Brave includes $5 of credits a month, and Firecrawl gives 1,000 credits a month. SerpApi's free plan covers 250 searches a month.
+Which search API does not depend on Google or Bing?
Brave Search API. Brave says every result comes from its own independent index, crawled and ranked by Brave, with no fallback to another search engine.
+How do I get Google search results in JSON for an AI app?
Use SerpApi. It parses Google Search, Google AI Mode, Google AI Overview and other Google result types into JSON, with plans from $25 a month for 1,000 searches and a free plan of 250 searches a month.
+What is the difference between Tavily and Exa?
Tavily bills in credits, with a basic search at 1 credit and pay as you go at $0.008 per credit, and bundles Search, Extract, Crawl, Map and Research. Exa bills per 1,000 requests from $4 and adds people, company and scholarly works indexes plus scheduled Monitors.
+Do search APIs return page content or only links?
It depends on the API. Exa, Perplexity Search API, Brave LLM Context and Jina's s.jina.ai return page text with results, while SerpApi returns parsed result pages with snippets and links, so you need a reader such as Firecrawl or Jina Reader for full content.
About this list
Updated November 12, 2026 by the Mojolin research desk. Mojolin is self-funded
Questions or corrections: editorial@mojolin.com
More best software lists