Claude API pricing is usage-based. You pay separately for input and output tokens, and the model you select usually determines most of the bill. In Anthropic’s pricing documentation retrieved in 2026, listed rates range from $0.25 per million input tokens for Claude Haiku 3 to $15 per million input tokens for Claude Opus 4 and 4.1; output tokens cost more than input tokens on every listed model.
For most production workloads, start with Haiku for routine high-volume work, Sonnet when quality and cost need to be balanced, and Opus only when its additional capability is worth the premium. Then reduce repeat-context costs with prompt caching, use Batch API for asynchronous jobs, and verify your organization’s live rate limits and model lifecycle status in the Anthropic Console before committing to an architecture.
As an Amazon Associate I earn from qualifying purchases.
How Claude API billing works
Anthropic meters Claude API usage by tokens. An input token is text, data, or instructions sent to the model; an output token is text generated in response. Each request is priced using the model’s input and output rates, so a long prompt and a long answer can cost very different amounts even when they are part of the same conversation.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
A practical estimate is:
cost = (input tokens ÷ 1,000,000 × input rate) + (output tokens ÷ 1,000,000 × output rate)
#1 Best Overall
- Compact & Portable Design: This Small Excel, Word, PowerPoint, & Windows Cheat Sheet Mouse Pad measures 11.8" x 9.8", offering a portable mouse pad. Its compact size fits in bags or laptop cases, suitable for professionals, students, or travelers.
- Multi-Software Shortcut Guide: Featuring essential shortcut keys for Excel, Word, PowerPoint, & Windows, this small mouse pad provides a quick reference for frequently used shortcuts, offering a reference designed to support workflow efficiency
- Smooth & Precise Surface: The finely-textured surface ensures precise mouse control, allowing for smooth and accurate movement during work or gaming sessions.
- Durable & Non-Slip Design: This Multi-Software Shortcut Mouse Pad features stitched edges to reduce fraying. The non-slip rubber base helps keep it securely in place, and the water-resistant fabric allows for easier maintenance.
- Clear Print Quality: Displays high-resolution printing intended to remain legible through regular use and cleaning
Your application may also incur separate tool charges. Anthropic’s web search tool is priced at $10 per 1,000 searches, in addition to normal token charges, according to Anthropic’s 2026 pricing documentation.
Current listed token rates
The following figures are the rates stated in Anthropic pricing documentation retrieved in 2026. They are per million tokens (MTok), not per request.
| Model | Input tokens | Output tokens | Typical fit |
|---|---|---|---|
| Claude Haiku 3 | $0.25/MTok | $1.25/MTok | Lowest-cost classification, extraction, and routing |
| Claude Haiku 3.5 | $0.80/MTok | $4/MTok | High-volume tasks needing more capability than Haiku 3 |
| Claude Sonnet 3.7 | $3/MTok | $15/MTok | General production workloads |
| Claude Sonnet 4 | $3/MTok | $15/MTok | General production workloads and stronger reasoning |
| Claude Opus 4 | $15/MTok | $75/MTok | Highest-capability use cases where quality justifies cost |
| Claude Opus 4.1 | $15/MTok | $75/MTok | Highest-capability use cases; verify current availability |
These are base token rates. They do not guarantee a particular response time, throughput allocation, context-window entitlement, or model lifespan. Model names and availability can change, so production systems should use the active model IDs shown in the current catalog.
Worked cost examples
Suppose a Sonnet request contains 100,000 input tokens and produces 10,000 output tokens. At $3/MTok input and $15/MTok output, the estimated request cost is $0.30 + $0.15, or $0.45.
The same token volumes on Opus 4.1 cost $1.50 for input plus $0.75 for output, totaling $2.25. On Haiku 3, they cost $0.025 plus $0.0125, totaling $0.0375. That difference is why model selection normally matters more than small prompt optimizations.
Rank #2
- [MULTIFUNCTIONAL]You'll get 2 pieces computer monitor memo boards that you can stick on the left and right edges of your monitor, and they're the perfect office desk organizers and accessories. Computer monitor side panels desktop organizer are suitable for home work or office,bringing convenience. Desktop memo is used to organize meeting memos, important messages, business cards, planning notes.Paste on the message board to keep track of important things and to-do items to prevent forgetting.
- [🌟HIGHLY QUALITY] The material of computer screen side note holder is transparent acrylic. Durable, simple, stylish, light weight, easy to use, not easy to fall off or break. This cute office supplies for women desk can be used for a long time. This computer desk accessories is waterproof and dirt resistance, and look simple and stylish. The transparent acrylic sticky note holder as cubicle accessories is easy to notice the context of your sticky notes.
- [📋Easy to use] Office must haves cool office gadgets for desk ready to tear, easy to install and remove, not easy to leave traces. You only need to peel off the protective film on the surface of the computer side board memo, wipe off the dust on the edge of the computer monitor, and then stick the desk essentials for women office on the right or left side of the tape, and you're done. A perfect gift for your colleagues, friends or classmates and family members or relatives
- [🏢MULTI-SCENE USE] This desk supplies computer memo board can be applied to home and office, clear your office decor for women, suitable for most computer monitors, screens and cabinets, you can put it where you think, this cute office decor serve as a reminder. Stick on the computer side. It’s a good office gadgets can remind work improve office productivity. Pasted cabinets, dressers, refrigerators, walls, etc as cubicle accessories. To make life more orderly.
- [💌NOTE] The adhesive force of the computer sticky note holder is very strong. It can not be directly pasted on the computer screen. It should pasted on the black edge of the screen. Narrow edge not recommended!!! If you are not satisfied with your purchase, or if the product is damaged or broken in transit, please let us know immediately. We will promptly solve your problem.
At scale, multiply the per-request estimate by actual traffic, but keep input and output volumes separate. A support bot that reads large documents but gives short answers is input-heavy; a code-generation system that returns long files is output-heavy and can be disproportionately expensive on Opus.
How prompt caching changes the economics
Prompt caching is intended for repeated context, such as a long system instruction, policy manual, product catalog, or codebase prefix that appears in many requests. Instead of paying the full input rate for every repeat, you pay a cache-write rate when content is stored and a much lower cache-read rate when that content is reused.
Cache write and read multipliers
| Operation | Price relative to base input rate | When it applies |
|---|---|---|
| Five-minute cache write | 1.25× | Context written with a five-minute lifetime |
| One-hour cache write | 2× | Context written with a one-hour lifetime |
| Cache read | 0.1× | Previously cached context reused before expiry |
For Sonnet at the $3/MTok base input rate, a five-minute write is $3.75/MTok, a one-hour write is $6/MTok, and a cache read is $0.30/MTok. The output rate is unchanged. Caching is therefore most useful when the same large prefix is read many times; it can cost more than ordinary input on the first write, but subsequent reads are heavily discounted.
When caching pays off
- Cache stable instructions and reference material, not a short prompt that changes on every request.
- Keep volatile user data after the cached prefix so a small change does not invalidate the whole context.
- Choose the five-minute lifetime for bursts and the one-hour lifetime when traffic is spread across a longer interval.
- Measure hit rate. A cache that rarely gets reused adds write charges without delivering read savings.
When the Batch API is cheaper
Anthropic’s Batch API processes requests asynchronously and applies a 50% discount to both input and output token prices. Anthropic states: “The Batch API allows asynchronous processing of large volumes of requests with a 50% discount on both input and output tokens.”
At that discount, a standard Sonnet request priced at $3/MTok input and $15/MTok output is effectively $1.50/MTok input and $7.50/MTok output for batch processing. A Haiku 3 request becomes $0.125/MTok input and $0.625/MTok output.
Rank #3
- Stay Organized with Ease: Our To Do List Notepad provides multiple sections with plenty of space to jot down all of your important tasks.A versatile office supply perfect for daily task management, helping you keep track of everything that needs to be done and prioritize effectively.
- Achieve Your Productivity Goals: Begin each day on a positive note, reminding you of your potential, strength, and the significance of staying focused on your goals. Our Daily Checklist Notepad is the ultimate tool to streamline your daily planning and elevate your productivity. With a clear and concise overview of your tasks, you can effortlessly prioritize what truly matters and make consistent strides towards achieving your goals.
- Reliable and Stylish: Our to do list planner combines functionality with durability. With a transparent PP cover, the inner pages are well-protected from dirt and wear. Additionally, the back cover is crafted with thick paperboard, offering a stable surface for your writing needs. Each notepad includes 52 pages of high-quality, 100gsm paper, which is thick and non-bleeding, resulting in a smooth and enjoyable writing experience.
- Versatile Use: This daily checklist notepad, suitable for teacher supplies and a wide range of applications, proves invaluable in various settings including work, school, and home. Whether you need to organize your daily tasks, plan a project, or create a grocery list, this notepad serves as the ideal tool. Its compact size ensures easy portability, allowing you to conveniently carry it with you throughout your day.
- Perfect Present for Anyone: Our Daily To-Do List Notepad is the perfect present for those who value organization and productivity. With its stylish design and versatility, this office desk accessory is also a practical addition to school supplies. Suitable for students, professionals, and homemakers alike, it helps keep tasks and priorities in check effortlessly.
Batch is a poor fit for interactive chat, checkout decisions, moderation that must block an action immediately, or any workflow with a strict response-time promise. It is a strong fit for overnight classification, document extraction, embedding-like labeling workflows, evaluation runs, backfills, and large report generation where completion can be asynchronous.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Why very long context can cost more
Long-context pricing is not always the same as the standard model rate. For the documented Sonnet 4 1M-context beta, inputs above 200,000 tokens are billed at $6/MTok and outputs at $22.50/MTok. Those rates apply to the specified long-context beta, not automatically to every Sonnet request.
Eligibility for that beta is restricted to usage tier 4 organizations or organizations with custom rate limits. If your workload can stay below 200,000 input tokens, splitting documents, retrieving only relevant passages, or summarizing earlier sections may avoid the higher bracket and reduce latency as well as cost.
Long-context budgeting checklist
- Count the complete serialized prompt, including system instructions, tool definitions, conversation history, and retrieved documents.
- Identify whether the request crosses the 200,000-token threshold rather than estimating from character count alone.
- Confirm that your organization is eligible for the 1M-context beta before designing around it.
- Price output separately; the beta output rate is $22.50/MTok, not the standard $15/MTok Sonnet rate.
Claude API rate limits: what you can and cannot assume
Anthropic uses account tiers rather than one universal requests-per-minute number. The documented structure includes Tier 1, Tier 2, Tier 3, and Tier 4, plus custom enterprise limits. Exact requests-per-minute and tokens-per-minute values depend on the organization, model, current capacity, and account configuration.
Do not copy a rate-limit number from another account into your capacity plan. Check the live Console and Anthropic’s current rate-limit documentation for your organization. Your application should still behave safely if limits are reached.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallCrashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteRank #4
- 【Adjustable & Ergonomic】：This laptop stand can be adjusted to a comfortable height and angle according to your actual needs, letting you fix posture and reduce your neck fatigue, back pain and eye strain. Very comfortable for working in home, office and outdoor.
- 【Sturdy & Protective】 ：Made of sturdy metal, it can support up to 17.6 lbs (8kg) weight on top; With 2 rubber mats on the hook and anti-skid silicone pads on top & bottom, it can secure your laptop in place and maximum protect your device from scratches and sliding. Moreover, smooth edges will never hurt your hands.
- 【Heat Dissipation】 ：The top of the laptop stand is designed with multiple ventilation holes. The open design offers greater ventilation and more airflow to cool your laptop during operation other than it just lays flat on the table.
- 【Portable & Foldable】：The foldable design allows you to easily slip it in your backpack. Ideal for people who travel for business a lot.
- 【Broad Compatibility】：Our desktop book stand is compatible with all laptops from 10-15.6 inches, such as MacBook Air/ Pro, Google Pixelbook, Dell XPS, HP, ASUS, Lenovo ThinkPad, Acer, Chromebook and Microsoft Surface, etc.Be your ideal companion in Home, Office & Outdoor.
Design for throttling
- Capture HTTP rate-limit responses and the server’s retry guidance.
- Retry with exponential backoff and jitter rather than issuing immediate parallel retries.
- Cap concurrency so a traffic spike does not create a retry storm.
- Queue non-urgent work and send it through Batch when an immediate response is unnecessary.
- Track both request volume and token volume; a few huge prompts can exhaust a token limit even when request count is low.
- Alert before the limit is reached so you can request a higher tier or redesign the workload.
Which Claude model should you choose?
Compare models on four required axes: capability and reasoning quality, input and output price, context-window needs, and operational limits or availability. For a production decision, add latency and migration risk.
Haiku: the cost-first default
Haiku is the logical starting point for high-volume classification, extraction, routing, tagging, and other routine operations. Haiku 3 has the lowest listed rates at $0.25/MTok input and $1.25/MTok output. Haiku 3.5 costs $0.80/MTok input and $4/MTok output and may be preferable when the cheaper model misses too many edge cases.
Use a representative evaluation set before switching an important workflow to the least expensive model. The cost saving is valuable only if rechecks, human review, or failed downstream actions do not erase it.
Sonnet: the balanced production choice
Sonnet 3.7 and Sonnet 4 are listed at $3/MTok input and $15/MTok output. That price-to-capability position makes Sonnet a sensible general default for customer support, document analysis, coding assistance, and multi-step business workflows when Haiku quality is insufficient and Opus is difficult to justify.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Use Sonnet 4’s 1M-context beta only when the long context is genuinely needed and your organization meets the tier or custom-limit requirement. Above 200,000 input tokens, apply the beta’s higher rates to the budget.
Best Value
- Twin-Ball Design : Hiacry gel pen features a double ball conical tip that reduces writing resistance and provides a stable and smooth writing experience. 0.7mm quick drying black ink ensures no jumping, leakage or seepage.
- Quick-Dry Long Lasting Ink：0.7mm black ink can slide effortlessly, dry immediately, & is not easily dirty. It has a high-capacity reservoir (up to 1300 meters), which is very suitable for left-handed & right-handed users.
- Sleek Design for Journaling & Planning: Engineered with a low center of gravity and precision tip for steady ink flow, these journaling pens excel at note-taking, sketching, planning, and more. They combine stylish pens with top-tier performance.
- Rolling Ball Design：The pocket clip with roll ball design can be easily attached to notebook pockets and binders, in addition, the fidget on the pen clip is also a small toy for your daily thinking to relieve stress.
- Comfortable Non-Slip Grip：The ergonomic pen barrel has a soft rubberized coating for a comfortable, non-slip grip, so you won't get fatigued even after long hours of writing.
Opus: pay for a measurable quality gain
Opus 4 and Opus 4.1 are listed at $15/MTok input and $75/MTok output, five times Sonnet’s input price and output price. Opus can make financial sense for difficult reasoning, complex code review, or high-value decisions where a quality improvement has a clear business benefit. It is hard to justify for routine extraction or large volumes of short, predictable tasks.
Because model names and lifecycle states change, re-check the current catalog before making Opus the long-term foundation of a system.
A simple selection test
- Run the same representative cases through Haiku, Sonnet, and Opus.
- Score accuracy, refusal behavior, formatting reliability, latency, and token usage.
- Estimate the monthly bill using separate input and output totals.
- Choose the least expensive model that meets your error and latency targets.
- Route exceptional cases to a stronger model instead of sending every request there.
Direct Claude API or Amazon Bedrock?
Both routes can provide access to Claude, but the better choice depends on procurement, security, and operational requirements rather than token price alone.
Recommended Free Tools
| Consideration | Direct Claude API | Amazon Bedrock |
|---|---|---|
| Billing | Anthropic-native usage billing | AWS billing and consolidated spend |
| Best fit | Teams wanting the simplest Anthropic integration and native features | Organizations standardized on AWS procurement, IAM, and regional routing |
| Regional endpoint pricing | Use Anthropic’s applicable API terms | Anthropic documents a 10% premium for applicable regional endpoints over global endpoints |
| Request payload limit | Use the direct API’s current limits | Anthropic documents a 20 MB Bedrock request payload limit |
| Operational choice | Direct access to Anthropic account controls | Integration with AWS governance and consolidated controls |
Choose the direct API when Anthropic-native features and straightforward billing are the priority. Consider Bedrock when your organization already buys AWS services, requires AWS identity and access management, needs consolidated billing, or needs regional routing. Compare the complete bill, including any regional premium and AWS-specific charges, rather than assuming the listed token rate tells the whole story.
Model retirement is part of the cost decision
Anthropic classifies models as active, legacy, deprecated, or retired. Requests to a retired model fail, so a low current price is not a durable bargain if the model is near the end of its lifecycle.
Before shipping, record the exact active model ID, monitor Anthropic’s lifecycle notices, and schedule migration work before the stated retirement date. Keep a tested fallback model and an evaluation suite so a migration does not become an emergency rewrite. This is especially important for applications embedded in customer workflows, where a failed model request can become an outage rather than a simple billing change.
Quick Recap
A practical cost-control workflow
- Measure tokens first. Log input and output token counts by model, feature, customer, and environment.
- Set a baseline. Calculate monthly cost from the formula and separate interactive traffic from asynchronous jobs.
- Route by difficulty. Send routine work to Haiku, balanced work to Sonnet, and only high-value exceptions to Opus.
- Cache stable context. Compare five-minute and one-hour write charges with expected read volume.
- Batch eligible work. Move non-urgent bulk processing to the Batch API for the 50% token discount.
- Control output length. Set appropriate maximum output limits and request concise structured formats when the task allows.
- Budget tool calls. Add web-search charges to token estimates when that tool is enabled.
- Protect the account. Use spend alerts, per-feature quotas, concurrency caps, and backoff handling.
- Review lifecycle status. Confirm that every production model ID is active and has a migration path.
Common pricing mistakes
- Using output rates for the whole request: input and output are metered separately.
- Ignoring prompt size: conversation history and tool schemas can make input tokens dominate.
- Assuming cache reads are free: they are discounted to 0.1× the base input rate, not zero.
- Applying the standard Sonnet rate to every context length: the documented Sonnet 4 beta charges higher rates above 200,000 input tokens.
- Using Batch for interactive traffic: the discount comes with asynchronous processing.
- Promising a universal requests-per-minute figure: limits are tiered and account-dependent.
- Choosing by model label alone: test quality, latency, context needs, availability, and migration risk together.
- Building on a retiring model: retired-model requests fail, so migration planning belongs in the initial design.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
