Tavily’s credit-based pricing can get confusing fast. Basic search costs 1 credit, advanced costs 2 credits, and operations like Crawl stack mapping and extraction costs together. Without understanding how credits work, you can burn through your monthly allocation faster than expected.

This guide breaks down Tavily’s pricing tiers, explains how credits actually work, and shows you where costs add up.

• Tavily uses credit-based pricing where operations cost 1-2 credits depending on depth (basic vs advanced)
• Plans range from free (1,000 credits) to $500/month (100,000 credits), plus $0.008 PAYG
• Credits don’t roll over monthly (unused credits disappear at month-end)
• Large gaps between tiers create pricing inefficiencies (4K to 15K to 38K to 100K credits)
• At 100K operations, Tavily costs $500-800
• Crawl operations stack costs (mapping + extraction), making budgeting unpredictable
• Best for search-first discovery; consider Firecrawl for structured extraction at scale

Tavily is aweb search API built specifically for AI agents and LLMs. It returns clean, structured results optimized for feeding directly into language models and RAG systems, rather than raw HTML meant for human browsing.

The platform uses a credit-based pricing model where different operations consume different amounts of credits.

Tavily offers four main endpoints: Search for discovering relevant pages, Extract for pulling content from specific URLs, Map for understanding website structure, and Crawl for navigating and extracting from entire sites.

Tavily offers a mix of monthly plans, pay-as-you-go pricing, and enterprise options. Understanding the credit system is crucial because different features consume different amounts of credits.

What you get:

• 1,000 credits per month
• Resets on the first day of each month
• No credit card required
• Access to all API endpoints (Search, Extract, Crawl, Map)

What it’s good for:

The free tier works for initial testing and prototyping. With 1,000 credits, you can make 1,000 basic searches or extract content from 5,000 URLs (at 1 credit per 5 URLs with basic extraction).

Limitations:

Credits don’t roll over. If you use 500 credits, the remaining 500 disappear at month-end. The 1,000 credit cap limits production testing. With advanced search (2 credits per request), you only get 500 searches. Advanced extraction reduces your capacity to 2,500 URLs.

What you get:

• 4,000 credits per month ($0.0075 per credit)
• Higher rate limits than free tier
• Monthly credit reset (unused credits don’t roll over)
• Full API access

What it’s good for:

Project suits MVPs or small applications with predictable, low-volume usage. At 4,000 credits, you can make 4,000 basic searches monthly (roughly 130 per day) or extract from 20,000 URLs with basic extraction.

Where it falls short:

With advanced search (2 credits each), your 4,000 credits become 2,000 searches (about 65 per day). Advanced extraction cuts your capacity to 10,000 URLs. That’s restrictive for growing applications.

At $30 monthly, this tier is nearly 2× the cost of Firecrawl’s Hobby plan ($16 for 3,000 pages), but with variable credit consumption that makes actual usage unpredictable.

Credits reset monthly with no rollover. Use 2,500 credits and lose 1,500. You either consistently max out usage or accept waste.

What you get:

• 15,000 credits per month ($0.0067 per credit)
• Better per-credit rate than Project
• Monthly credit reset (unused credits don’t roll over)
• Full API access

What it’s good for:

Bootstrap suits growing projects that need more capacity than Project but aren’t ready for Startup-level volumes. At 15,000 credits, you can make 15,000 basic searches monthly (500 per day) or extract from 75,000 URLs with basic extraction.

Where it falls short:

At $100 monthly, you’re getting 15,000 credits versus Firecrawl’s Standard plan at $83 for 100,000 pages. That’s 6.7× fewer operations for $17 more.

The gap between Project (4,000 credits for $30) and Bootstrap (15,000 credits for $100) is steep. If you need 8,000-10,000 credits monthly, you’re forced to either pay PAYG rates or jump to Bootstrap and potentially waste credits.

What you get:

• 38,000 credits per month ($0.0058 per credit)
• Better per-credit rate than Bootstrap
• Monthly credit reset (unused credits don’t roll over)
• Full API access

What it’s good for:

Startup works for established applications with consistent mid-volume usage. At 38,000 credits, you can make 38,000 basic searches monthly (roughly 1,265 per day) or extract from 190,000 URLs with basic extraction.

Where it falls short:

At $220 monthly for 38,000 credits, the cost per operation still exceeds alternatives. Firecrawl’s Standard plan gives you 100,000 pages for $83 (2.6× more capacity at 38% of the cost).

The pricing ladder creates awkward gaps. If you need 50,000 credits, you’re forced to either pay Bootstrap + PAYG overage or jump to Growth ($500 for 100,000 credits) and waste 50,000 credits.

What you get:

• 100,000 credits per month ($0.005 per credit)
• Best per-credit rate on fixed plans
• Higher rate limits
• Priority support

What it’s good for:

Growth makes sense for established applications with consistent, high-volume usage. The $0.005 per credit rate beats both PAYG ($0.008) and lower tiers. If you’re consistently using 70,000+ credits monthly, this tier offers better economics than PAYG rates.

Where it falls short:

At $500 monthly for 100,000 credits, you’re paying 6× more than Firecrawl’s Standard plan ($83 for 100,000 pages). The cost difference is substantial for extraction-heavy workflows.

With advanced search (2 credits each), your 100,000 credits become 50,000 searches. With advanced extraction, you can process 250,000 URLs. While that’s high volume, Firecrawl delivers the same 100,000 pages with consistent features at a fraction of the cost.

Credits still reset monthly with no rollover. Use 85,000 credits and lose 15,000 ($75 worth). This creates pressure to maximize usage or accept waste.

What you get:

• No monthly commitment
• Pay only for what you use
• $0.008 per credit
• Scales with your usage

What it’s good for:

PAYG works for applications with unpredictable or bursty traffic patterns. If usage varies significantly month-to-month (seasonal projects, ad-hoc research), you avoid paying for unused credits.

Where it falls short:

At $0.008 per credit, PAYG is 60% more expensive than Growth’s $0.005 per credit. For 100,000 credits (100K basic searches), you’re paying $800 via PAYG. Compare that to Firecrawl’s Standard plan at $83 for 100,000 pages. That’s nearly 10× more expensive.

The cost compounds quickly. At 50,000 credits monthly, you’re paying $400 versus Firecrawl’s $83. The flexibility comes at a significant premium.

What you get:

• Custom credit allocation
• Negotiated per-credit rates
• Dedicated support
• Custom rate limits
• SLAs and contractual guarantees

What it’s good for:

Enterprise suits large organizations with millions of API calls monthly, strict compliance requirements, or dedicated infrastructure needs. If you’re processing 1M+ credits monthly, custom pricing negotiations can lower your effective per-credit cost below the $0.005 Growth rate.

Where it falls short:

You have to contact sales for pricing, creating friction if you want to evaluate costs quickly. There’s no transparency into what Enterprise pricing looks like until you’re in sales conversations.

Tavily uses a credit-based system where different operations and features consume different amounts of credits. Understanding this structure is essential for budgeting accurately.

Your search depth determines the cost:

• Basic search: 1 credit per request
• Advanced search: 2 credits per request

Basic search works for straightforward queries where you need quick results. Advanced search provides deeper, more comprehensive results but doubles the cost.

The number of successful URL extractions and your extraction depth determine the cost. You’re only charged for successful extractions.

• Basic extract: 1 credit per 5 successful URLs
• Advanced extract: 2 credits per 5 successful URLs

Example: Extract content from 50 URLs with basic depth = 10 credits. The same 50 URLs with advanced depth = 20 credits.

The number of pages mapped and whether you use natural language instructions determines the cost.

• Regular mapping: 1 credit per 10 pages mapped
• Mapping with instructions: 2 credits per 10 pages mapped

Example: Map a 100-page website without instructions = 10 credits. With instructions = 20 credits.

Crawl combines both mapping and extraction, so costs stack:

Crawl Cost = Mapping Cost + Extraction Cost

Example scenarios:

• Crawl 10 pages with basic extraction = 1 credit (mapping) + 2 credits (extraction) = 3 total credits
• Crawl 10 pages with advanced extraction = 1 credit (mapping) + 4 credits (extraction) = 5 total credits
• Crawl 50 pages with basic extraction = 5 credits (mapping) + 10 credits (extraction) = 15 total credits

Research follows dynamic pricing with minimum and maximum boundaries per request:

The actual cost depends on query complexity, sources accessed, and depth of research performed. You won’t know the exact cost until after the request completes, making budgeting difficult.

Unlike flat-rate pricing where one page always equals one credit, Tavily’s variable consumption means:

• A basic search costs 1 credit, but an advanced search costs 2 credits (same operation, double the cost)
• Extracting 100 URLs could cost 20 credits (basic) or 40 credits (advanced)
• Crawling 100 pages could cost 30 credits or 50+ credits depending on extraction depth
• Research queries could cost anywhere from 4 to 250 credits per request

This variability makes it challenging to predict your monthly bill. If your application’s requirements change (you start using advanced search more often, or add instructions to mapping operations), your credit consumption can double without changing the number of requests you make.

Tavily’s credit-based pricing creates budgeting challenges that drive developers to explore alternatives.

The same operation costs differently based on depth. Basic search costs 1 credit, advanced costs 2 credits. Research API ranges from 4 to 250 credits per request, with no way to predict the final cost until after completion.

This variability makes forecasting monthly bills nearly impossible when your application’s requirements change.

Tavily’s tier jumps create painful choices:

• Project: 4,000 credits ($30) → Bootstrap: 15,000 credits ($100)
• Bootstrap: 15,000 credits ($100) → Startup: 38,000 credits ($220)
• Startup: 38,000 credits ($220) → Growth: 100,000 credits ($500)

Need 25,000 credits monthly? You’re forced to pay Bootstrap plus PAYG overage, or jump to Startup and waste 13,000 credits. Need 60,000 credits? No good option exists between Startup and Growth.

At 100,000 operations, the cost difference is stark:

• Tavily Growth: $500 for 100,000 credits
• Tavily PAYG: $800 for 100,000 credits
• Firecrawl Standard: $83 for 100,000 pages

That’s 6× to 10× more expensive for similar volumes.

Crawl combines mapping and extraction costs. Crawl 100 pages with basic extraction costs 30 credits. With advanced extraction, that jumps to 50 credits for the same 100 pages. This stacking isn’t obvious until you’re already using the API.

Choosing between Firecrawl and Tavily comes down to what you’re actually building.

Choose Firecrawl when:

• You need structured data extraction. Firecrawl is purpose-built for pulling clean, structured data from websites. Use natural language prompts to define what you need, no CSS selectors required. Perfect for product catalogs, pricing monitoring, financial data, or competitive intelligence.
• Cost matters at scale. At 100K pages monthly, Firecrawl costs $83 versus Tavily’s $500-800. The 6-10× cost difference compounds quickly for production workloads.
• You’re dealing with complex sites. JavaScript-heavy SPAs, sites requiring authentication, paginated content, forms, CAPTCHAs. Firecrawl’s Agent handles these automatically at no extra charge. Tavily treats them as edge cases with variable credit consumption.
• Predictable pricing is important. One page equals one credit, always. No multipliers, no variable consumption, no surprise bills.

Read our detailed comparison between Tavily and Firecrawl along with other Tavily alternatives here.

Try Firecrawl free (500 credits, no card required) or read the docs to see how extraction-first architecture handles your use case.

How much does Tavily cost per month?

Tavily offers plans from free (1,000 credits) to $500/month (100,000 credits). Project costs $30 for 4,000 credits, Bootstrap costs $100 for 15,000 credits, Startup costs $220 for 38,000 credits. Pay-as-you-go costs $0.008 per credit with no monthly commitment.

Do Tavily credits roll over to the next month?

No. Tavily credits reset on the first day of each month. Unused credits disappear and don’t carry forward. If you purchase 15,000 credits and only use 10,000, you lose the remaining 5,000 at month-end.

Why does Tavily cost more than alternatives like Firecrawl?

Tavily charges based on operation depth and features. At 100K operations, Tavily costs $500-800 versus Firecrawl’s $83. Tavily’s variable credit consumption (basic vs advanced, with multipliers for features) makes it significantly more expensive at scale than flat-rate alternatives.

What’s the difference between basic and advanced search in Tavily?

Basic search costs 1 credit and provides standard results. Advanced search costs 2 credits (double the price) and delivers deeper, more comprehensive results. The same pattern applies to extraction: basic costs 1 credit per 5 URLs, advanced costs 2 credits.

How do Tavily Crawl operations affect my credit usage?

Crawl combines mapping and extraction costs, which stack together. Crawling 100 pages with basic extraction costs 30 credits total (10 for mapping + 20 for extraction). With advanced extraction, the same 100 pages costs 50 credits.