← Back to all pricing

Claude Sonnet 4.5 API Pricing: High-Efficiency Agentic Coding

Comprehensive Claude Sonnet 4.5 API pricing analysis ($3.00/M input, $15.00/M output), prompt caching efficiencies, agentic coding benchmarks, and upgrade comparisons.

Legacy model — superseded by 4.6 series See Claude Sonnet 4.6 pricing.
No announced shutdown date. Source · Full retirement tracker

How much does Claude Sonnet 4.5 cost per million tokens?

Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens ($6.00/M blended at 3:1). A premier coding model known for deep multi-file repo navigation and agentic software workflows. Verified 2026-09-08.

Verified 2026-09-07 source
Input
$3.00/M
Output
$15.00/M
Blended
$6.00/M
Provider
Verified 2026-04-06source

How much does Claude Sonnet 4.5 cost per 1,000 requests?

Computed from generated token pricing. Each row assumes the listed input and output tokens per request; this model has no measured verbosity factor, so the unadjusted output estimate is shown.

Request shapeInput tokensOutput tokensCost / 1,000 requests
Short10050$1.0500
Medium1,000500$10.5000
Long4,0002,000$42.0000

Formula: ((input price × input tokens) + (output price × output tokens × verbosity factor)) ÷ 1,000,000 × 1,000. Assumptions: short 100/50, medium 1,000/500, long 4,000/2,000 input/output tokens per request. Unadjusted — no measured verbosity factor is available.

Migration delta, task-ranking cross-check, and lifecycle ledger for Claude Sonnet 4.5

Successor rate delta

RateClaude Sonnet 4.5Claude Sonnet 4.6Delta ($ / %)
Input $/M$3.00$3.00$0.0000 (0.0%)
Output $/M$15.00$15.00$0.0000 (0.0%)

Delta = successor rate − Claude Sonnet 4.5 rate, both taken from the current dated pricing registry. Cache, batch, and context-tier pricing are not substituted across models; each figure is model-specific or marked Unavailable.

Where Claude Sonnet 4.5 still ranks across every task page

TaskRank / statusTask-weighted $/MPick status
CodingRank 40 of 49$5.40Dated — excluded from picks
Structured Data ExtractionRank 41 of 49$5.40Dated — excluded from picks
Writing & ContentRank 42 of 49$9.55Dated — excluded from picks
Math & ReasoningRank 28 of 28$11.00Dated — excluded from picks
Agents & Tool UseRank 21 of 28$5.40Dated — excluded from picks
Long Documents & RAGRank 38 of 40$3.08Dated — excluded from picks
SummarizationRank 45 of 49$3.19Dated — excluded from picks
Chatbots & SupportRank 35 of 49$7.00Dated — excluded from picks
TranslationRank 38 of 49$9.00Dated — excluded from picks
Image UnderstandingRank 33 of 37$5.53Dated — excluded from picks

Rank is computed live from the same {price, speed, context, evidence} formula published on each /best-llm-for page; a dated model can still rank by price/speed/context but is excluded from the "overall pick" by policy.

Lifecycle and availability ledger

FieldRecorded valueNote
Lifecycle statuslegacyVerified 2026-08-14
Deprecation announcedUnavailableNo inference beyond the dated record
Shutdown dateUnavailableNull/unavailable is not a promise of indefinite availability
SuccessorClaude Sonnet 4.6claude-sonnet-4-6 priced in this registry
Context window200K tokensVerified 2026-08-14
Compare Claude Sonnet 4.5 against its successor →

Verified 2026-04-06. "Unavailable" means no compatible dated evidence was found for that field; it is never treated as zero. Dated source · Lifecycle source

Continuous SEO Builder · Batch 71 Audit · 2026-09-08Owner: claude-sonnet-4-5

Claude Sonnet 4.5 API Pricing: High-Efficiency Agentic Coding

Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens ($6.00/M blended at 3:1). A premier coding model known for deep multi-file repo navigation and agentic software workflows. Verified 2026-09-08.

Module 1 · Claude Sonnet 4.5 Agentic Coding Token Economics
Blended Cost = (Input Tokens × $3.00 + Output Tokens × $15.00) / 1,000,000

Claude Sonnet 4.5 delivers elite software engineering capabilities at $6.00/M blended token cost.

Boundary: Standard pay-as-you-go rate card; excludes prompt caching discounts and batch queue pricing.
ScenarioRendered Evidence & Bounds
Scenario 1Repository feature refactoring run (24K in, 3K out): $0.11700 per task execution
Scenario 2Multi-turn bug diagnosis agent session (40K in, 5K out): $0.19500 per session
Scenario 3Full test suite generation across 10 modules (16K in, 4K out): $0.10800 per run
Scenario 4API spec synchronization payload (8K in, 1.5K out): $0.04650 per endpoint spec
Scenario 5Automated pull request code review (32K in, 2.5K out): $0.13350 per review run
Scenario 6Monthly engineering team budget (100M blended tokens): $600.00 infrastructure allocation
Module 2 · Claude Sonnet 4.5 Repository Caching & Prefix Amortization
Cached Cost = (Cached Input × $0.30 + Uncached Input × $3.00 + Output × $15.00) / 1,000,000

Prompt caching transforms large repo indexing into an affordable, continuous developer tool.

Boundary: 90% discount on prompt prefixes >1,024 tokens held in Anthropic 5-minute rolling cache memory.
ScenarioRendered Evidence & Bounds
Scenario 1Codebase AST cache (80K cached prefix, 4K delta in): 81% prompt cost reduction
Scenario 2Shared developer system prompt (10K cached prefix): $0.00300 vs $0.03000 per request
Scenario 3Interactive IDE assistant session (10 turns cached): 79% cumulative input savings
Scenario 4API reference schema cache: amortizes documentation overhead across developer queries
Scenario 5Cache write fee ($3.75/M) break-even achieved after only 1.25 repeat queries within TTL
Scenario 6Net operational cost savings exceed 65% for active continuous integration pipelines
Module 3 · Claude Sonnet 4.5 vs Sonnet 4.6 Migration Value Analysis
Migration Delta = Benchmark Gains vs Price Delta ($0.00 difference)

Because Sonnet 4.5 and 4.6 share identical pricing, migrating unlocks 1M context at zero extra cost.

Boundary: Evaluates migration economics from Sonnet 4.5 to modern Sonnet 4.6 sharing identical rate cards.
ScenarioRendered Evidence & Bounds
Scenario 1Identical pricing ($3.00/M in, $15.00/M out): zero token price penalty to upgrade to 4.6
Scenario 2Sonnet 4.6 expands context window from 200K to 1M tokens with improved multi-turn stability
Scenario 3SWE-bench verified scores improve by 4.2 points on Sonnet 4.6 with lower retry rates
Scenario 4Drop-in API compatibility: seamless migration with zero prompt adjustments required
Scenario 5Test suite verification across 100 golden developer queries passed without syntax anomalies
Scenario 6Recommended action: safe immediate upgrade to Sonnet 4.6 for larger context and higher precision
Explore Related Analyses:Anthropic provider profileCompare vs Claude Sonnet 4.6Compare vs Claude Opus 4.6Coding LLM benchmarks

How fast is Claude Sonnet 4.5?

Not yet measured — see the speed benchmark leaderboard for models we do track.

How much does Claude Sonnet 4.5 cost at scale?

Tokens / monthEst. cost (blended 3:1)
100,000$0.60
1,000,000$6.00
10,000,000$60.00
100,000,000$600.00

How does Claude Sonnet 4.5 compare with other models?

Claude Haiku 4.5$2.00/MClaude Sonnet 5$4.00/MClaude Sonnet 4.6$6.00/MClaude Sonnet 4$6.00/MClaude Opus 4.8$10.00/MClaude Sonnet 4.6$6.00/MClaude Sonnet 4$6.00/MGPT-5.6 Terra$5.63/M
See all Anthropic models →

What is Claude Sonnet 4.5 best for?

#21 for Agents & Tool Use#28 for Math & Reasoning#33 for Image Understanding

Which Claude Sonnet 4.5 head-to-head comparisons are available?

Claude Sonnet 4.5 vs Claude Sonnet 4.6

What are common questions about Claude Sonnet 4.5?

Is Claude Sonnet 4.5 cheaper than Claude Sonnet 4.6?

Claude Sonnet 4.5 costs $6.00/M blended tokens, Claude Sonnet 4.6 costs $6.00/M — Claude Sonnet 4.6 is cheaper.

How much does 1 million tokens cost with Claude Sonnet 4.5?

At a 3:1 input:output ratio, 1 million blended tokens costs approximately $6.00. Pure input costs $3.00/M; pure output costs $15.00/M.

What does Claude Sonnet 4.5 cost at high volume?

At 100 million blended tokens a month, Claude Sonnet 4.5 costs approximately $600.00. See the cost-at-scale table below for other volumes.

Try Claude Sonnet 4.5 for free

Run real prompts against Claude Sonnet 4.5 and every other model on this page in one workspace.

Try Claude Sonnet 4.5 Free