RETIRES NOV 30 2026DEPRECATED200K CONTEXTTEXT + VISIONBATCH -50%
Claude Sonnet 4.5 API Pricing
Claude Sonnet 4.5 is deprecated. Anthropic announced on September 30, 2026 that it retires from the Claude API on November 30, 2026, and names Claude Sonnet 5.5 as the replacement. Until then it still bills at $3.00/M input, $15.00/M output and $0.30/M cache-hit input. Checked by hand against platform.claude.com.
Input - per 1M tokens
$3.00/M
Stable Anthropic table flat
Output - per 1M tokens
$15.00/M
Stable since launch flat
Cached input - 90% off
$0.30/M
Cache 5min or 1h -90%
Effective - agentic blend
$2.05/M
92/8 split - 82% cache
§ 01 / TERMINAL
Run the numbers.
Calculator pre-loaded with Claude Sonnet 4.5 rates, for the months left before the November 30, 2026 retirement and for sizing the move to Sonnet 5.5. Tweak spend, output mix, or cache hit rate; share the URL to share the calculation.
$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
§ 02 / SCENARIOS
Real-world presets.
CODING AGENT
Repo-wide bug fix
$0.180/task
LONG DOC ANALYSIS
Reading 100-page contracts
$0.281/doc
RAG SUPPORT
Support agent ticket triage
$0.019/ticket
ASSISTANT
Research planning turn
$0.041/turn
Estimate · anthropic-bpe-estimate · ≈3.5 chars/token Auto-counts as you type
This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.
Characters 341
Words 54
Tokens (estimated) 97 tokens
Cost as input · uncached $0.00029 USD
Cost as output · uncached $0.00145 USD
Cost as cached input $0.00003 USD
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| Claude Sonnet 5.5 | $2.00 cache $0.10 | $10.00 | $1.29 −37% | 1M | Anthropic's named replacement |
| Claude Opus 4.7 | $5.00 cache $0.50 | $25.00 | $3.41 +67% | 1M | Frontier reasoning and hard code |
| Claude Opus 4.6 | $5.00 cache $0.50 | $25.00 | $3.41 +67% | 1M | Frontier reasoning and hard code |
| Claude Opus 4.5 | $5.00 cache $0.50 | $25.00 | $3.41 +67% | 200K | Frontier reasoning and hard code |
| Claude Opus 4.1 | $15.00 cache $1.50 | $75.00 | $10.24 retired | 200K | Legacy Opus workloads |
| Claude Sonnet 4.6 | $3.00 cache $0.30 | $15.00 | $2.05 same | 1M | Production agents and coding |
| Claude Sonnet 4.5 Current | $3.00 cache $0.30 | $15.00 | $2.05 agentic 92/8 | 200K | Production agents and coding |
| Claude Haiku 4.5 | $1.00 cache $0.10 | $5.00 | $0.682 −3.0× | 200K | Support and classification |
| GPT-5.4 | $2.50 cache $0.25 | $15.00 | $1.80 −12% | 1.05M | Tool use and app agents |
| Gemini 3.8 Flash | $0.75 cache $0.075 | $3.75 | $0.481 −4.3× · promo | 1M | Long-context document analysis |
Frequently asked.
Practical pricing questions, with vendor list prices separated from workload assumptions.
Q · 01 What is Claude Sonnet 4.5 priced at? +
Anthropic's pricing page lists
$3.00/M input, $15.00/M output, and $0.30/M cache-hit input for Claude Sonnet 4.5, unchanged since launch and unchanged by the deprecation. Under the 92/8 agentic blend with 82% cache hits, that works out to $2.05/M.Q · 02 When is Claude Sonnet 4.5 retired? +
Anthropic deprecated
claude-sonnet-4-5-20250929 on September 30, 2026 and retires it from the Claude API on November 30, 2026. Until then it keeps answering at the rates on this page. Anthropic defines a retired model as no longer available for use: requests to it fail. The recommended replacement is Claude Sonnet 5.5 (claude-sonnet-5-5).Q · 03 Will moving to Sonnet 5.5 lower the bill? +
Per token, yes: Sonnet 5.5 lists at
$2.00/M input and $10.00/M output, against $3.00 and $15.00 here. Per task, by less than that suggests. Anthropic's migration guide says Sonnet 5.5's tokenizer produces about 30% more tokens than Sonnet 4.5's for the same text, so recount tokens on your own prompts before budgeting the gap. Under this site's blend the per-token figures are $1.29/M against $2.05/M.Q · 04 What breaks when switching to Sonnet 5.5? +
Anthropic's migration guide lists the changes for code written against Sonnet 4.5: a prefilled last assistant turn returns a 400 error; thinking budgets give way to an effort level, and thinking is on when no
thinking field is sent; non-default temperature, top_p and top_k must go; and tool-call input should be read with a standard JSON parser. Set output_config.effort explicitly and re-run cost baselines after the switch.Q · 05 How does prompt caching change the price? +
Cache hits are listed at
$0.3/M, which is 10% of the $3/M base input rate. Anthropic also charges cache writes at $3.75/M for 5-minute writes and $6/M for 1-hour writes. The calculator models repeated prompt sections as cache hits.Q · 06 Is there a Batch API discount? +
Yes. Anthropic's Batch API gives a 50% discount on input and output tokens. For Claude Sonnet 4.5, the batch table maps to
$1.5/M input and $7.5/M output.Q · 07 Does long context cost extra? +
This page uses the model's documented
200K context label. Longer-context Opus 4.6, Opus 4.7, and Sonnet 4.6 use standard pricing across the full 1M window. Prompt caching and batch discounts still apply according to their normal rules.Q · 08 Does regional pricing differ? +
For first-party Claude API US-only inference, Anthropic applies a
1.1x multiplier to all token categories for this generation. Bedrock and Vertex AI publish their own regional pricing policies.Q · 09 Are volume discounts published? +
No fixed public volume-discount ladder is published. Enterprise and committed-spend contracts can have private terms, but standard API users should assume the list prices shown here unless they have a separate agreement.
Q · 10 How accurate is the tokenizer estimate? +
The live counter uses a Claude-family English estimate of
4.875 characters per token. Actual billing uses Anthropic's server-side tokenizer; code, tables, and non-English text can differ materially.