Claude Opus 5.5 API Pricing / Price per 1M tokens
Anthropic • API Price Comparison
Also known as: anthropic-claude-opus-5.5 · anthropic claude opus 5.5 · anthropicclaudeopus5.5 · anthropic-claude-opus5.5 · anthropic claude opus5.5 · claude-opus-5.5 · claudeopus5.5 · claude-opus5.5 · claude opus5.5 · opus-5.5 · opus 5.5 · opus5.5
API pricing and token costs at a glance
Updated on · Published currencies / per 1M tokens; cheapest ranked by input
| Channel | Input / 1M | Output / 1M | Cache read / 1M | Updated / verified on |
|---|---|---|---|---|
| AnthropicOfficial | USD 4 | USD 20 | USD 0.2 | 2026-11-06 |
| XiuRouterLowest input | USD 0.92 | USD 4.599999 | USD 0.046 | 2026-11-06 |
| OpenRouter | USD 4 | USD 20 | USD 0.2 | 2026-11-06 |
Metered API cost = input tokens / 1M × input rate + output tokens / 1M × output rate. At the lowest-input-rate channel, 1M input plus 250K output tokens costs approximately USD 2.07. This excludes caching, tools, retries and other service fees; the lowest input rate does not guarantee the lowest total bill.
— means the price or date is not recorded. Cross-currency ranking uses exchange rates.
Claude Opus 5.5 is a large language model by Anthropic with a 1M-token context window. Across the 3 API channels tracked here, input pricing ranges from $0.92–$4.00 per 1M input tokens (USD-normalized); the cheapest is XiuRouter at $0.92 per 1M input tokens. The official channel (Anthropic) charges $4.00 per 1M input tokens and $20.00 per 1M output tokens. It is also available through subscription plans such as Claude Free, Claude Enterprise, Copilot Pro, Copilot Standalone. Its Agent Arena net-improvement score is approximately 13.8%. Prices are collected daily from official provider pages and every tracked channel, then cross-checked by a data-accuracy audit.
Claude Pro · Opus 5.5 quota (x2v measurement)
Measured on:
| Window | Output-heavy | Agentic coding (99% cache reads, 0.2% output) |
|---|---|---|
| 5 hours | ≈1.3M output tokens, $26–29 | ≈160M total tokens, $44 |
| 1 week (≈7.2 five-hour windows, 6.8–7.8) | 9.3M output tokens, $185–205 | 1.15B tokens, $320 |
| 1 month (weekly cap ×4.35) | 40M output tokens, $800–890 | 5.0B tokens, $1,390 |
x2v first-party measurement (2026-11-04, Claude Code); not an official Anthropic quota.
147 controlled Opus 5.5 calls via headless `claude -p --model claude-opus-5-5 --effort medium`, no tools/MCP; list prices $4/M input, $20/M output, $8/M 1h cache writes, $0.20/M cache reads.
API-equivalent value is ≈40–70× the $20/mo Pro price.
Meter fit: cache writes ≈0.55×, cache reads ≈0.5× list; L ≈ $25.75 output-equivalent per 5h. Thinking tokens count as output. The weekly cap binds.
Limits: integer-percent meter (±0.5pt), single day/time, one account, batch B2 possibly throttled.
Model performance benchmarks
Scores retain their original benchmark, version and metric. We do not combine them into a site-wide score; results from different versions are not directly comparable.
Coding agent performance
Artificial Analysis Coding Agent Index: per-task performance, API cost and runtime when this model runs inside different agents (Claude Code, Codex, …). Each configuration is evaluated separately.
This page is in the top-10 traffic review set. Each key price and entitlement shows its source, conditions, last verification, and whether it is expired or an estimate. Estimates are not confirmed entitlements.
💳 Subscription Plans
Plans from Anthropic come first, followed by linked third-party plans; each group is organized by product line.
Official plans
These are the model producer’s plans. A plan may not include this exact model; check the official site for current model availability and allowances.
💬 Chat subscriptions
7 plansMonthly access to the assistant apps and their model line-up.
- Plan price: $0.00/month
- Payment methods: Credit card / Debit card / Apple Pay / Google Pay
Conditions and details
- Currency USD, unit per month
- Published free tier
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Does the subscription allowance cover third-party models?: Unknown
- Estimate~20 messages/day limit.
- Web access only.
- Contact sales / Request a quote
- Payment methods: Invoice / Contract
Conditions and details
- Currency USD, unit per month
- Contact sales; a zero or blank price is not a public purchase price
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Does the subscription allowance cover third-party models?: Unknown
- Contact sales.
- EstimateReported floor ~$50k/yr, ~50-70 seat minimum.
- SCIM, SSO, audit, 500K context.
- Plan price: $20.00/month
- Billed annually $200.00
- Payment methods: Credit card / Debit card / Apple Pay / Google Pay
Conditions and details
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Does the subscription allowance cover third-party models?: Unknown
- Monthly API credit (USD): 20
- Plan price: $25.00/month
- Billed annually $240.00 ($20.00/month)
Conditions and details
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Does the subscription allowance cover third-party models?: Unknown
- Includes $20/month Claude API credit per Standard seat; all seats pool into one team balance capped at $500/month (discounted Team plans same).
- Covers the Claude API, Claude Managed Agents, the Claude Agent SDK and playground in the Claude Console only; excludes Claude Code, extra usage in the Claude apps, and Claude on AWS/Bedrock/Google Cloud/Microsoft Foundry.
- Claim by linking a Console organization after 7 days on the plan.
- Refreshes each billing cycle (monthly on annual plans); unused credit expires, no rollover.
- Standard seat, United States, before tax.
- Monthly $25; annual $20/seat/month ($240/seat/year).
- Minimum 2 seats (teams of 2 to 150).
- Includes Claude Code.
- verified 2026-10-19.
- and and
- 5x Pro usage every 5 hours and every week.
- Fable included, up to half the weekly usage limit.
- $100/month in API credits to build on the Claude Platform.
- Monthly billing only; no annual plan.
Conditions and details
- Claude Code included; shared usage limits apply.
- Monthly API credit (USD): 100
- Plan price: $100.00/month
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Does the subscription allowance cover third-party models?: Unknown
- Usage per week: 5× claude-team
- Monthly API credit (USD): 100
- Plan price: $125.00/month
- Billed annually $1,200.00 ($100.00/month)
Conditions and details
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Does the subscription allowance cover third-party models?: Unknown
- Includes $100/month Claude API credit per Premium seat; all seats pool into one team balance capped at $500/month (discounted Team plans same).
- Covers the Claude API, Claude Managed Agents, the Claude Agent SDK and playground in the Claude Console only; excludes Claude Code, extra usage in the Claude apps, and Claude on AWS/Bedrock/Google Cloud/Microsoft Foundry.
- Claim by linking a Console organization after 7 days on the plan.
- Refreshes each billing cycle (monthly on annual plans); unused credit expires, no rollover.
- Premium seat, United States, before tax.
- Monthly $125; annual $100/seat/month ($1,200/seat/year).
- Minimum 2 seats (teams of 2 to 150).
- Includes Claude Code; 5x more usage than a Standard seat.
- verified 2026-10-19.
- and and
- 10x Pro usage every week.
- No 5-hour limit for verified accounts.
- Fable included, up to half the weekly usage limit.
- $200/month in API credits to build on the Claude Platform.
Conditions and details
- Monthly billing only; no annual plan.
- Claude Code included; shared usage limits apply.
- Monthly API credit (USD): 200
- Plan price: $200.00/month
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Does the subscription allowance cover third-party models?: Unknown
Third-party plans
⌨️ Coding plans
5 plansEditor and CLI coding agents, metered in prompts rather than tokens.
- Copilot Pro: $10 USD per month.
- $15 total monthly AI credits = $10 base + $5 flex (1,500 credits).
- Interactions on Free, Pro and Pro+ may be used for training unless opted out.
- This is not a discount in exchange for data.
Conditions and details
- HydraFusion and HydraFusion Max: research preview; enable experimental/preview features.
- Model pool: Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna.
- Billing multiplier undisclosed.
- Model links describe the preview pool, not direct model selection or guaranteed inclusion in credits.
- Recorded plan allowance: 1500 credits per month
- Included credits: 1500
- Included usage: 1500 credits
- Plan price: $10.00/month
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Quota unit credits
- Does the subscription allowance cover third-party models?: Unknown
- GitHub AI Credits: 1 credit = $0.01.
- Flex credits may change.
- Annual pricing, regional access and payment methods.
- Copilot Standalone: $10 USD per user per month.
- Per-user monthly seat fee only; no included AI credits; metered usage billed separately from first use at model rates against the enterprise budget.
- Sales-assisted only through a GitHub/Microsoft account team from 2026-10-28.
- Self-serve is not yet available (coming in stages).
Conditions and details
- Unavailable to GitHub Free/Team organizations and self-service GitHub Enterprise Cloud trials.
- Same features and model eligibility as Copilot Enterprise, subject to administrator policies and spending controls.
- Code completion and next edit suggestions work with no budget.
- Copilot Chat, code review and Copilot cloud agent require available enterprise budget.
- An exhausted budget never removes the seat.
- One billing model per enterprise: Copilot access (GitHub Enterprise Cloud-included or Standalone) cannot coexist with Business/Enterprise seats in the same enterprise.
- Transition is voluntary.
- Copilot access through GitHub Enterprise Cloud: $0 Copilot seat charge for users who hold a GitHub Enterprise Cloud license (license cost excluded here).
- Same features/models as Copilot Enterprise; no included AI credits; metered usage billed separately from first use against the enterprise budget.
- Sales-assisted only for now.
- Enterprise owners must still assign seats; this option can coexist with Standalone.
- Standalone training policy; the supplied Business/Enterprise training statement does not explicitly name Standalone.
- HydraFusion and HydraFusion Max: research preview; enable experimental/preview features.
- Model pool: Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna.
- Billing multiplier undisclosed.
- Model links describe the preview pool, not direct model selection or guaranteed inclusion in credits.
- Included credits: 0
- Included usage: 0 credits
- Plan price: $10.00/month
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Quota unit credits
- Does the subscription allowance cover third-party models?: Unknown
- GitHub AI Credits: 1 credit = $0.01.
- Flex credits may change.
- Annual pricing, regional access and payment methods.
- Copilot Business: $19 USD per user per month.
- 1,900 AI credits per user per month, pooled across the organization.
- Business and Enterprise interactions are not used for training.
- HydraFusion and HydraFusion Max: research preview; enable experimental/preview features.
Conditions and details
- Model pool: Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna.
- Billing multiplier undisclosed.
- Model links describe the preview pool, not direct model selection or guaranteed inclusion in credits.
- Recorded plan allowance: 1900 credits per month
- Included credits: 1900
- Included usage: 1900 credits
- Plan price: $19.00/month
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Quota unit credits
- Does the subscription allowance cover third-party models?: Unknown
- GitHub AI Credits: 1 credit = $0.01.
- Flex credits may change.
- Annual pricing, regional access and payment methods.
- Copilot Pro+: $39 USD per month.
- $70 total monthly AI credits = $39 base + $31 flex (7,000 credits).
- Interactions on Free, Pro and Pro+ may be used for training unless opted out.
- This is not a discount in exchange for data.
Conditions and details
- HydraFusion and HydraFusion Max: research preview; enable experimental/preview features.
- Model pool: Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna.
- Billing multiplier undisclosed.
- Model links describe the preview pool, not direct model selection or guaranteed inclusion in credits.
- Recorded plan allowance: 7000 credits per month
- Included credits: 7000
- Included usage: 7000 credits
- Plan price: $39.00/month
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Quota unit credits
- Does the subscription allowance cover third-party models?: Unknown
- GitHub AI Credits: 1 credit = $0.01.
- Flex credits may change.
- Annual pricing, regional access and payment methods.
- Copilot Enterprise: $39 USD per user per month.
- 3,900 AI credits per user per month, pooled across the organization.
- Available for organizations using GitHub Enterprise Cloud.
- Business and Enterprise interactions are not used for training.
Conditions and details
- HydraFusion and HydraFusion Max: research preview; enable experimental/preview features.
- Model pool: Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna.
- Billing multiplier undisclosed.
- Model links describe the preview pool, not direct model selection or guaranteed inclusion in credits.
- Recorded plan allowance: 3900 credits per month
- Included credits: 3900
- Included usage: 3900 credits
- Plan price: $39.00/month
- Currency USD, unit per month
- EstimateFeatures, message caps, context and rate limits come from the same plan row; lines marked Estimate are not guaranteed allowances
- Quota unit credits
- Does the subscription allowance cover third-party models?: Unknown
- GitHub AI Credits: 1 credit = $0.01.
- Flex credits may change.
- Annual pricing, regional access and payment methods.
Which plan is right for you?
💬 Chat subscriptions: Monthly access to the assistant apps and their model line-up.
⌨️ Coding plans: Editor and CLI coding agents, metered in prompts rather than tokens.
| Channel | Type | Input / 1M | Output / 1M | Rate Limit | vs Official | ||
|---|---|---|---|---|---|---|---|
XiuRouter💰 Best Price | Reseller | $0.92 | $4.60 | - | -77.0% | ||
OpenRouter13 variants | Aggregator | $4.00 | $20.00 | - | - | ||
Show all price variants for this channelOpenRouter Auto / default routeDefault $4.00 / $20.00 per 1M default_route Cheapest endpoint: Amazon Bedrock $4.00 / $20.00 per 1M cheapest_endpoint Azure | anthropic/claude-opus-5.5-20260921 $4.00 / $20.00 per 1M upstream_provider Google | anthropic/claude-opus-5.5-20260921 $4.00 / $20.00 per 1M upstream_provider Anthropic | anthropic/claude-opus-5.5-20260921 $4.00 / $20.00 per 1M upstream_provider Amazon Bedrock | anthropic/claude-opus-5.5-20260921 $4.00 / $20.00 per 1M upstream_provider Claude Platform on AWS | anthropic/claude-opus-5.5-20260921 $4.00 / $20.00 per 1M upstream_provider Google | anthropic/claude-opus-5.5-20260921 $4.40 / $22.00 per 1M upstream_provider Amazon Bedrock | anthropic/claude-opus-5.5-20260921 $4.40 / $22.00 per 1M upstream_provider Amazon Bedrock | anthropic/claude-opus-5.5-20260921 $4.40 / $22.00 per 1M upstream_provider Azure | anthropic/claude-opus-5.5-20260921 $4.40 / $22.00 per 1M upstream_provider Google | anthropic/claude-opus-5.5-20260921 $4.40 / $22.00 per 1M upstream_provider Anthropic | anthropic/claude-opus-5.5-20260921 $8.00 / $40.00 per 1M upstream_provider | |||||||
Anthropic2 variants🏢 Official | Official | $4.00 | $20.00 | - | Baseline | ||
Show all price variants for this channelStandardDefault $4.00 / $20.00 per 1M tier Fast mode $8.00 / $40.00 per 1M tier | |||||||
| Usage Level | Tokens/Month | XiuRouter | OpenRouter | Anthropic |
|---|---|---|---|---|
| 🐣 Light | 0.1M in + 0.1M out | $0.32 | $1.40 | $1.40 |
| 📊 Medium | 1.0M in + 0.5M out | $3.22 | $14.00 | $14.00 |
| 🚀 Heavy | 10.0M in + 5.0M out | $32.20 | $140.00 | $140.00 |
| 🏢 Enterprise | 100.0M in + 50.0M out | $322.00 | $1,400.00 | $1,400.00 |
Frequently asked questions about claude-opus-5.5
How much does the Claude Opus 5.5 API cost?
The lowest input rate is USD 0.92 per 1M tokens; output on that channel is USD 4.599999 per 1M tokens.
Which channel has the cheapest Claude Opus 5.5 API?
XiuRouter has the lowest input rate: USD 0.92 per 1M tokens. Cross-currency ranking uses exchange rates; compare output and cache prices separately.
What is the Claude Opus 5.5 cache-read price?
XiuRouter: USD 0.046; OpenRouter: USD 0.2; Anthropic: USD 0.2 per 1M cache-read tokens.
When was Claude Opus 5.5 pricing last updated?
Updated on 2026-11-06. This is the latest recorded price update or verification; see the table for individual channel dates.
Is there a free Claude Opus 5.5 API?
No available recorded channel lists both input and output at zero. Missing prices, free cache reads or a free chat tier do not establish a free API; check provider terms for trial credits.
How do I calculate Claude Opus 5.5 token costs and monthly spend?
Metered API cost = input tokens / 1M × input rate + output tokens / 1M × output rate. At the lowest-input-rate channel, 1M input plus 250K output tokens costs approximately USD 2.07. This excludes caching, tools, retries and other service fees; the lowest input rate does not guarantee the lowest total bill.
How much Opus 5.5 does Claude Pro include?
Measured on: 2026-11-04. 5 hours: Output-heavy ≈1.3M output tokens, $26–29; Agentic coding (99% cache reads, 0.2% output) ≈160M total tokens, $44. 1 week (≈7.2 five-hour windows, 6.8–7.8): Output-heavy 9.3M output tokens, $185–205; Agentic coding (99% cache reads, 0.2% output) 1.15B tokens, $320. 1 month (weekly cap ×4.35): Output-heavy 40M output tokens, $800–890; Agentic coding (99% cache reads, 0.2% output) 5.0B tokens, $1,390. x2v first-party measurement (2026-11-04, Claude Code); not an official Anthropic quota. 147 controlled Opus 5.5 calls via headless `claude -p --model claude-opus-5-5 --effort medium`, no tools/MCP; list prices $4/M input, $20/M output, $8/M 1h cache writes, $0.20/M cache reads. API-equivalent value is ≈40–70× the $20/mo Pro price. Meter fit: cache writes ≈0.55×, cache reads ≈0.5× list; L ≈ $25.75 output-equivalent per 5h. Thinking tokens count as output. The weekly cap binds. Limits: integer-percent meter (±0.5pt), single day/time, one account, batch B2 possibly throttled. https://claude.ai/artifact/DcK5UbxR1SmkgDn38QtBo2
Related pricing research