Updated September 1, 2026 with Claude Fable 5.1
Token Estimator – paste text to estimate tokens and cost
Model Comparison – see costs across all models
| Model | Per Request | 10K Requests |
|---|
Calculating Claude API Costs
Anthropic charges per million tokens, with separate rates for input and output. Total Claude API costs depend on which model you select, how much of your context repeats across requests, and whether your workload can run asynchronously. This page covers current pricing as of September 1, 2026, including Claude Fable 5.1, the features that reduce spend, and the considerations that matter when choosing a model.
Paying for API Access
The Claude API is pay as you go and billed on usage. There is no subscription required to call it, and no monthly minimum. An API key itself costs nothing to create, and holding one costs nothing; you are billed only for the tokens your requests actually consume. New organizations receive a small amount of free credit to test with, after which requests need a funded balance.
API billing is separate from a Claude Pro or Max subscription. Paying for a subscription does not include API credit, and API spend does not draw down a subscription. If your usage is interactive rather than programmatic, a subscription is usually the cheaper route; the API earns its place when you are calling Claude from your own software.
Current Claude Models
Anthropic released Claude Fable 5.1 on September 1, 2026 at $10 input and $50 output per million tokens, the same base rates as Fable 5. The change is in cache reads: a cache hit on Fable 5.1 costs $0.25 per million tokens, 2.5% of the input price, where every other Claude model charges 10%. Anthropic estimates that cuts typical Fable workloads by around 25% and highly agentic, cache-heavy workloads by up to around 45%. Claude Mythos 5.1 is the same model with more permissive safeguards, priced identically and available only through Anthropic’s trusted access programs. Claude Opus 5 remains the model Anthropic suggests starting with for most workloads at $5 / $25, and the default on Claude Max. Claude Sonnet 5’s $2 / $10 launch price is now its standard price; the increase to $3 / $15 that had been scheduled for September 1 will not happen. Haiku 4.5 is the lowest-cost current-generation option.
Anthropic’s most advanced generally available model, released September 1, 2026. Use the API identifier claude-fable-5-1. Base pricing matches Fable 5, but cache reads cost $0.25 per million tokens instead of $1.00, so any workload that reuses a long prefix is cheaper on 5.1 at the same headline rate. Cache writes are unchanged at $12.50 (5-minute) and $20 (1-hour), and the minimum cacheable prompt is 512 tokens. Default effort is high; low and medium effort match or beat Fable 5 at a lower cost per task. Available on the Claude API, Amazon Bedrock, Google Cloud and Microsoft Foundry.
Identical to Fable 5.1 in weights and price, including the $0.25 cache read, with more permissive safeguards for cybersecurity and life-sciences work. Access is through the Cyber Verification Program and the Life Sciences Verification Program, currently limited to a set of US organizations. Claude Mythos 5, the counterpart to Fable 5, remains on the same limited-access terms at Fable 5 rates.
The current Opus model, priced identically to Opus 4.8 on every line including cache writes, cache reads and batch. Use the API identifier claude-opus-5. Two changes affect cost planning: the minimum cacheable prompt drops to 512 tokens, half that of Opus 4.8, and thinking runs by default so total output tokens include thinking. The effort setting from low to max controls how many tokens Opus 5 spends, which changes the bill through token volume rather than through the rate.
The previous Fable generation, superseded by Fable 5.1 on September 1, 2026 at the same $10 / $50 base rate. Use the API identifier claude-fable-5. The cost difference between the two is entirely in cache reads: $1.00 per million tokens here against $0.25 on Fable 5.1. Fable 5 ships with safety classifiers that can decline certain requests; a refused request that produces no output is not billed, and the API can automatically retry on another Claude model via the fallbacks parameter.
The Opus flagship for high-difficulty coding, complex agentic work, and expensive-to-fail business tasks at half the Fable 5 rate. Use the API identifier claude-opus-4-8. Fast mode is available for Opus 4.8 at 2x standard token pricing when speed is worth the premium.
Released June 30, 2026 and the best starting point for most API workloads, with agentic performance approaching Opus 4.8 at a lower price. Use the API identifier claude-sonnet-5. It launched at $2 input and $10 output per million tokens as introductory pricing, and on September 1, 2026 Anthropic confirmed that as the standard price; the scheduled increase to $3 / $15 will not occur. Sonnet 5 uses an updated tokenizer that can map the same text to roughly 1.0 to 1.35x more tokens, so re-profile prompts when migrating from Sonnet 4.6.
Superseded by Sonnet 5, which now costs less at $2 / $10. Sonnet 4.6 stays available at $3 / $15 for workloads pinned to its behavior or to the older tokenizer.
The cheapest current-generation model. Best for routing, extraction, summarization, moderation, real-time chat, and sub-agent work under a Sonnet or Opus orchestrator.
Full Claude API Pricing Table
Standard prices per million tokens, ordered by model family: Opus, Sonnet, then Haiku. Prompt caching and batch discounts apply at standard rates unless a feature says otherwise. Retired models are included only for historical comparison.
| Model | Input | Output | Status |
|---|---|---|---|
| Fable family – frontier capability | |||
| Claude Fable 5.1 | $10.00 | $50.00 | Active; newest; 1M context; cache reads $0.25 |
| Claude Mythos 5.1 | $10.00 | $50.00 | Limited availability; same model as Fable 5.1; cache reads $0.25 |
| Claude Fable 5 | $10.00 | $50.00 | Active; previous Fable generation; cache reads $1.00 |
| Claude Mythos 5 | $10.00 | $50.00 | Limited availability via Project Glasswing; 1M context |
| Opus family – flagship workhorse | |||
| Claude Opus 5 | $5.00 | $25.00 | Active; current Opus; 1M context; fast mode 2x; caches from 512 tokens |
| Claude Opus 4.8 | $5.00 | $25.00 | Active; 1M context; fast mode 2x |
| Claude Opus 4.7 | $5.00 | $25.00 | Active; 1M context; fast mode withdrawn 24 Jul 2026 |
| Claude Opus 4.6 | $5.00 | $25.00 | Active; 1M context; fast mode discontinued 29 Jun 2026 |
| Claude Opus 4.5 | $5.00 | $25.00 | Active; 200K context |
| Claude Opus 4.1 | $15.00 | $75.00 | Active; higher legacy Opus price |
| Claude Opus 4 | $15.00 | $75.00 | Deprecated; retires June 15, 2026 |
| Sonnet family – balanced | |||
| Claude Sonnet 5 | $2.00 | $10.00 | Active; 1M context; launch price made standard Sep 1, 2026 |
| Claude Sonnet 4.6 | $3.00 | $15.00 | Active; previous Sonnet; 1M context |
| Claude Sonnet 4.5 | $3.00 | $15.00 | Active; 200K context |
| Claude Sonnet 4 | $3.00 | $15.00 | Deprecated; retires June 15, 2026 |
| Claude 3.7 Sonnet | $3.00 | $15.00 | Retired on Claude API; historical only |
| Haiku family – fastest and lowest cost | |||
| Claude Haiku 4.5 | $1.00 | $5.00 | Active; current Haiku; 200K context |
| Claude Haiku 3.5 | $0.80 | $4.00 | Retired on Claude API; historical only |
| Claude Haiku 3 | $0.25 | $1.25 | Retired on Claude API; historical only |
Claude Mythos 5 shares Fable 5’s capabilities and pricing but is offered only in limited release to approved customers through Anthropic’s Project Glasswing, so Fable 5 is the generally available model in that class. Both carry 30-day data retention and are not available under zero data retention agreements, which matters for compliance-sensitive workloads. Opus 4.1 and Opus 4 at $15/$75 are now three times more expensive than Opus 4.8 for weaker performance. Teams still running on them can reduce costs materially by migrating to a current Opus model without a sticker-price increase over Opus 4.5 through 4.7. Haiku 3 at $0.25/$1.25 remains the absolute cheapest option for simple tasks.
Opus Fast Mode Costs
Fast mode runs Opus at around 2.5x the default speed for 2x standard token pricing, which is $10 per million input tokens and $50 per million output tokens. It is available on Opus 5 and Opus 4.8, on the first-party Claude API only. Opus 4.6 fast mode was discontinued on June 29, 2026, and Opus 4.7 fast mode was removed on July 24, 2026, so the 6x rate that applied to it no longer exists.
What this means A standard Opus 5 request with 10,000 input tokens and 2,000 output tokens costs $0.10. The same request in fast mode costs $0.20. Use fast mode when response time is the bottleneck, not as a default setting for batch jobs or background processing.
Prompt caching still matters in fast mode because repeated input can be billed at cache-read rates after the fast-mode multiplier. Batch processing is separate: fast mode is not available with the Batch API, so non-urgent work should usually use batch instead of paying the speed premium. Fast mode is an Opus feature; Anthropic does not list it for Fable 5, Sonnet, or Haiku models.
Prompt Caching
Prompt caching is the single most effective cost feature on the Claude API. When a request reads from the cache, cached tokens are billed at 10% of the standard input rate, a 90% discount. On Claude Fable 5.1 and Claude Mythos 5.1 the read rate is 2.5% of the input price, $0.25 per million tokens, a 97.5% discount. Cache writes cost 1.25x the standard input rate for a 5-minute time-to-live, or 2x for a 1-hour time-to-live, on every model.
The break-even math is straightforward. A 5-minute cache pays for itself after a single read. A 1-hour cache pays for itself after two reads. Every subsequent read saves 90% of the normal input cost, or 97.5% on Fable 5.1 and Mythos 5.1. The pay-back point is the same on those two models; the saving per read is larger.
Minimum cacheable prompt length
A prompt has to reach a minimum length before it can be cached at all, and that minimum varies by model. Fable 5.1, Mythos 5.1, Opus 5, Fable 5 and Mythos 5 cache from 512 tokens. Opus 4.8, Sonnet 5, Sonnet 4.6 and Sonnet 4.5 need 1,024. Opus 4.7 and Haiku 3.5 need 2,048. Haiku 4.5, Opus 4.6 and Opus 4.5 need 4,096, the highest floor of any current model.
Nothing fails loudly when a prompt falls short. The request is processed without caching and no error is returned, so the saving simply does not appear on the bill. Check the response usage fields to confirm: if both cache_creation_input_tokens and cache_read_input_tokens come back as 0, the prompt was not cached. If a prompt falls just short of the threshold, padding the cached section to reach it is usually worth doing.
What to cache
Cache any context that repeats across requests. System prompts, role definitions, tool schemas, reference documents, code examples, and long context that does not change between user messages are all candidates. A chatbot with a 4,000-token system prompt serving 10,000 conversations per day will save roughly $108 daily on Sonnet 4.6 by caching that system prompt, which is about $3,240 per month from a single optimization.
Which TTL to use
Use the 5-minute TTL for interactive workloads where the same context is hit in quick succession, such as chat sessions, multi-turn agents, and iterative coding. Use the 1-hour TTL when cache hits are spread out but predictable within an hour, or when combining cache with batch processing.
Batch Processing
The Message Batches API processes requests asynchronously and returns results within 24 hours, typically under 1 hour, at exactly 50% off both input and output prices. There is no quality difference between batch and real-time responses. The only trade-off is timing.
Batch fits any workload where a few hours of latency is acceptable: document processing pipelines, data enrichment, nightly analytics, offline evaluations, content generation queues, code reviews, and dataset classification. A team processing 500,000 documents per month at Sonnet 4.6 rates saves approximately $750 to $2,250 monthly by routing through batch.
Combining batch with caching
Batch processing and prompt caching compose. A cached read inside a batch request is billed at 50% of the 10% cache-read rate, which works out to 5% of the standard input rate. On prompt-heavy workloads with repeated context, combined use can reduce input costs by up to 95%. Pair the 1-hour cache TTL with batch rather than the 5-minute TTL, since batch turnaround usually exceeds 5 minutes.
Choosing a Claude Model
Model selection has the largest impact on cost. For cost-sensitive production workloads, start on Sonnet 5 and only move up when benchmarks on your specific workload show a meaningful quality gap. Opus 5 covers most complex agentic coding and expensive-to-fail work; Anthropic’s own guidance is to start there for most workloads. Fable 5.1 sits above it at double the base rate for the most demanding reasoning and long-horizon agents, and its $0.25 cache reads narrow that gap on context-heavy work.
When to use Fable 5.1
Use Fable 5.1 when Opus 5 at higher effort still falls short: the hardest multi-step reasoning, agents that run unattended for hours, and research or analysis where capability is worth a 2x base premium. Two things change the cost picture against Fable 5. Cache reads are $0.25 per million tokens, so a workload that reuses a large prefix on most requests brings its effective input rate close to Opus 5’s, although output stays at 2x; run the cache planner above with your real prefix size and read count to see the blended rate. Effort is the other lever: Anthropic’s launch data shows low and medium effort matching or beating Fable 5 at a fraction of the tokens, so start at medium and raise it only where your evals show the difference. Adaptive thinking is always on and bills at output rates. Plan for refusal handling; a refused request that produces no output is free.
When to use Opus 5
Opus 5 is the default choice for deep reasoning, long-horizon agents and agentic coding, at the same $5 / $25 as Opus 4.8. Because thinking is on by default, budget for output tokens that include the model’s thinking, and revisit any max_tokens value carried over from a workload that ran without thinking. Lower effort settings produce strong results at a fraction of the tokens, so effort is the first control to reach for when cost matters more than depth.
When to use Sonnet 5
Sonnet 5 suits most production workloads: agentic coding, tool use, computer use, and everyday knowledge work. It is the most agentic Sonnet to date and lands close to Opus 4.8 on many tasks, at $2 / $10, a third less than Sonnet 4.6 on both input and output. Move up to Opus 5 or Fable 5.1 only when benchmarks on your own workload show a clear quality gap worth the higher token price.
When to stay on Fable 5
Fable 5 costs the same as Fable 5.1 on base tokens and more on cache reads, and Fable 5.1 scores higher on every benchmark Anthropic published, so the only reason to keep traffic on Fable 5 is a workload you have validated against its specific behavior and have not yet re-evaluated. Pin the version, run your evals on Fable 5.1, and migrate once they pass.
When to use Opus 4.8
Use Opus 4.8 when the task requires sustained reasoning across many steps, when errors are costly, or when the model must recover from its own mistakes during long autonomous runs. Complex debugging, large-scale code migration, long-horizon agents that manage spreadsheets or documents end-to-end, and enterprise workflows where a wrong answer is expensive are the strongest fits.
When Haiku 4.5 is enough
Haiku 4.5 handles classification, extraction, routing, summarization, moderation, and simple chat competently. At $0.10 per million tokens on cache hits, it is the most cost-effective choice for retrieval pipelines with heavily reused context. Haiku also works as a sub-agent under an Opus or Sonnet orchestrator in multi-agent architectures, where the orchestrator plans and Haiku executes.
Output tokens cost more than input
Output is billed at five times the input rate across every current model. A request with 5,000 input tokens and 5,000 output tokens on Sonnet 4.6 costs $0.015 for the input and $0.075 for the output, so the output represents 83% of the bill. Keep responses concise. Request specific output formats such as JSON or bullet points when the task allows. Set length constraints when detailed analysis is not required. Extended thinking, while billed at input rates, can add substantial volume on complex tasks, so monitor thinking token usage on long-running agents.
Claude Fable 5.1 Benchmarks
Scores from Anthropic’s Fable 5.1 announcement on September 1, 2026, with Fable 5 and Opus 5 re-run under the same conditions. Fable 5.1 was evaluated with its production safeguards on; where a safeguard intervened, the task scored zero, so these are floor figures for the model. Mythos 5.1 is the same model without those interventions and scored 60.9% on Terminal-Bench 4.0 against Fable 5.1’s 55.8%. OSWorld 2.0 uses the August 2026 task release, which is why no competitor score is shown there.
Claude Fable 5 Benchmarks
Benchmark comparison from Anthropic’s Fable 5 announcement. Anthropic reports Fable 5 and Mythos 5 scores jointly; on starred benchmarks Fable 5 can land closer to Opus 4.8 because its safety fallbacks route some queries away.
Claude Sonnet 5 Benchmarks
Benchmark comparison from Anthropic’s Claude Sonnet 5 launch on June 30, 2026. Sonnet 5 sits clearly above Sonnet 4.6 across every headline benchmark and approaches Opus 4.8, edging ahead of the flagship on GDPval-AA v2 knowledge work. Opus 4.8 is shown as the reference ceiling.
Additional API Charges
Base token costs are not the only line item on an API bill. Several features carry their own pricing.
Web search
$10 per 1,000 searches for the search tool itself. The input and output tokens produced by processing the search results are billed separately at the model’s standard rate.
Code execution
Code execution is free when the request also includes the web search or web fetch tool. Used on its own it is billed by execution time: each organization receives 1,550 free container-hours per month, and usage beyond that runs at $0.05 per hour per container, with a 5-minute minimum per execution.
Managed Agents sessions
$0.08 per session-hour of active runtime. Idle time while the session waits for the next message does not count. Standard token rates still apply on top.
US-only data residency
For workloads that must run inside the US, specifying US-only inference via the inference_geo parameter adds a 1.1x multiplier across every token category, covering input, output, cache writes, and cache reads. Applies to Claude 4.6 and later, so Fable 5.1, Mythos 5.1, Fable 5, Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, Sonnet 5 and Sonnet 4.6, on the first-party Claude API and Claude Platform on AWS. Earlier models return an error if the parameter is sent. Third-party platforms have their own regional pricing.
Tool use overhead
Tool use adds a small system prompt automatically, and the size depends on the model: about 286 to 354 tokens on Opus 5, Opus 4.8 and Sonnet 5, about 496 to 589 on the 4.5 and 4.6 models, and 675 to 804 on Opus 4.7. The text editor tool adds around 700 tokens, and bash adds 325 on Opus 5 and 4.8 or 244 on Opus 4.6 and earlier. The overhead is per-request, so it matters most on high-volume, short-message workloads.
1M Context Window
Fable 5.1, Mythos 5.1, Fable 5, Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, Sonnet 5, and Sonnet 4.6 include the full 1 million token context window at standard flat pricing. A 900,000-token request is billed at the same per-token rate as a 9,000-token request. This is a change from earlier generations, where long-context requests above 200,000 tokens incurred a 2x premium on all tokens in the request.
Prompt caching and batch discounts apply at standard rates across the full context window. For applications that analyze large codebases, long legal documents, or entire books in a single pass, the flat pricing keeps spend predictable.
Planning Token Counts
Claude’s tokenizer produces roughly 1.33 tokens per word for English prose, or about 4 characters per token. A planning rule of thumb: 1,000 words is approximately 1,330 tokens. Technical documentation runs slightly higher at around 1,400 tokens per 1,000 words. Source code runs higher still at around 1,500 tokens per 1,000 words, because punctuation and syntax tokenize less efficiently than prose.
For accurate counts before deployment, the Anthropic API includes a token counting endpoint that reports exact token usage for any input. The estimator above applies the 4-characters-per-token heuristic, which is close enough for budgeting but not exact. Re-run token counts when migrating between Opus releases because identical text can produce different billable token totals across model versions.
Best Cost Control Practices
A few operational practices keep API spend predictable as usage grows.
Track token usage per request from the start. The API response includes a usage object reporting exact input and output token counts. Monitor per-feature and per-user spend rather than only the monthly total, since the total alone will not identify which feature or customer is responsible for a spike.
Cache anything that repeats. System prompts, tool schemas, documentation embedded in the prompt, and few-shot examples are all candidates. Any context that appears in more than one request should be cached.
Route non-urgent requests through batch. Anything that does not need a sub-second response, such as nightly jobs, content generation pipelines, and evaluation runs, belongs on the Batch API at half price.
Set spending limits in the Anthropic Console. Organization-level and workspace-level limits prevent runaway spend from a bug or a misconfigured agent. Alerts at 50%, 75%, and 90% of the monthly budget give time to react before hitting the hard cap.
Benchmark smaller models on your actual task. Teams often default to Opus when Sonnet or even Haiku would suffice. Running a representative sample of real workload through each tier and measuring the quality gap empirically is usually worth the evaluation time, because the cost difference is large.
Pricing verified from Anthropic’s pricing documentation.
See also the detailed pricing documentation and model overview.
