Start free

Claude · Mid-tier · 200K context window

Claude Sonnet 5 pricing for math reasoning

$0.0330 per call at 1,500 input / 3,000 output tokens.

Per call

$0.0330

10K calls/mo

$330

Cache savings

90% off input

Batch savings

50% off all

Calculate your Claude Sonnet 5 costs

Adjust token counts, volume, and discount toggles to model your math reasoning spend. Pricing updated weekly.

Configure your usage

1,500
100100K
3,000
5050K
10,000
1001M

Estimated cost

Monthly cost
$330
10,000 calls/month
Cost per call$0.033
Daily cost$11
Input rate$2.00/MTok
Output rate$10.00/MTok
Analyze your real spend

Quick start

Call Claude Sonnet 5 with the Anthropic Python SDK:

from anthropic import Anthropic

client = Anthropic()
response = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=3000,
    messages=[{"role": "user", "content": "Your prompt here"}]
)
# Cost: ~$0.0330 per call at standard rates
# Input:  $2.00/MTok
# Output: $10.00/MTok

How Claude Sonnet 5 compares on cost

Per-call cost for all models scoring 50+ on math reasoning benchmarks. Hover for details.

Gemini 3.5 Flash-Lite$0.004875Gemini 3 Flash$0.009750Claude Haiku 4.5$0.0165GPT 5.6 Luna$0.0195Gemini 3.5 Flash$0.0292Claude Sonnet 5$0.0330Gemini 3.1 Pro$0.0390GPT 5.4$0.0488GPT 5.6 Terra$0.0488Claude Sonnet 4.6$0.0495Claude Opus 4.8$0.0825GPT 5.5$0.0975GPT 5.6 Sol$0.0975Claude Fable 5$0.1650

Cheapest viable option: Gemini 3.1 Pro at $0.0390/call (-18% less, 90/100 quality).

Complete Claude Sonnet 5 pricing breakdown

Standard token rates

Direction$/MTokTokens per call
Input$2.001,500
Output$10.003,000

Prompt caching: 90% off input tokens

Caching drops input cost from $2.00 to $0.20/MTok. For repeated system prompts, that saves $0.002700 per call.

Batch API: 50% off all tokens

Submit requests asynchronously and receive 50% off all token costs. This works best for offline pipelines where latency requirements are flexible.

Monthly cost projections

Monthly volumeCost per callStandard monthlyCached monthlyYou save with caching
10,000 calls$0.0330$330$303$27
50,000 calls$0.0330$1,650$1,515$135
100,000 calls$0.0330$3,300$3,030$270
500,000 calls$0.0330$16,500$15,150$1,350
1,000,000 calls$0.0330$33,000$30,300$2,700

Claude Sonnet 5 quality benchmarks for math reasoning

Use Claude Sonnet 5 for math reasoning

Scores 90/100 (rank #2 of 14). Strong pick for production math reasoning workloads.

Top-ranked: Claude Fable 5 (98/100) at $0.1650/call

Performance across all task types

Claude Sonnet 5 scores by task type. Math Reasoning highlighted.

Code Review96Code Generation93Summarization91Reasoning *90Q&A89Extraction88Creative Writing87Classification86
Task typeConfidence scoreRank out of 14
Code Review96/100#1
Code Generation93/100#2
Summarization91/100#1
Reasoning(current page)90/100#2
Q&A89/100#2
Extraction88/100#2
Creative Writing87/100#3
Classification86/100#3

Alternative models for math reasoning

8 models score 80+ on math reasoning benchmarks, sorted by input rate. Cheapest: Gemini 3.1 Pro (-18% less, 90/100 quality).

ModelProviderInput $/MTokOutput $/MTokQualityPer callvs Claude
Gemini 3.1 ProGemini$2.00$12.0090/100$0.039018% more
GPT 5.4GPT$2.50$15.0080/100$0.048848% more
GPT 5.6 TerraGPT$2.50$15.0082/100$0.048848% more
Claude Sonnet 4.6Claude$3.00$15.0088/100$0.049550% more
Claude Opus 4.8Claude$5.00$25.0097/100$0.0825150% more
GPT 5.5GPT$5.00$30.0095/100$0.0975195% more
GPT 5.6 SolGPT$5.00$30.0096/100$0.0975195% more
Claude Fable 5Claude$10.00$50.0098/100$0.1650400% more

About Claude Sonnet 5

Provider

Claude

Tier

Mid-tier

Context

200K tokens

Released

Jun 2026

Introductory pricing $2/$10 through Aug 31, 2026 (registry reflects the intro rate; standard $3/$15 resumes Sep 1 — tracked revert: KNOWN_ISSUES #34)

Other models from Claude

ModelTierInput $/MTokOutput $/MTokContext
Claude Opus 4.8Flagship$5.00$25.00200K
Claude Sonnet 4.6Mid-tier$3.00$15.00200K
Claude Haiku 4.5Fast / Budget$1.00$5.00200K
Claude Fable 5Flagship$10.00$50.001,000K

You can compare Claude Sonnet 5 head-to-head against any other model on our model comparison pages, or track how pricing has changed over time on the pricing history tracker.

Claude Sonnet 5 pricing for other use cases

See how Claude Sonnet 5 costs change across different workload types.

Code Generation

$0.0440/call

Unit Test Generation

$0.0560/call

SQL Generation

$0.0130/call

API Generation

$0.0660/call

Code Refactoring

$0.0480/call

Code Review

$0.0300/call

PR Review

$0.0460/call

Security Review

$0.0370/call

Bug Detection

$0.0300/call

Document Summarization

$0.0300/call

PDF Summarization

$0.0450/call

Meeting Notes

$0.0360/call

Email Summarization

$0.0110/call

Chatbot

$0.0140/call

RAG Pipeline

$0.0270/call

Customer Support Bot

$0.0160/call

FAQ Bot

$0.0110/call

Data Extraction

$0.0180/call

JSON Extraction

$0.0140/call

Invoice Parsing

$0.010000/call

Resume Parsing

$0.0130/call

Chain of Thought

$0.0560/call

Legal Analysis

$0.0600/call

Financial Analysis

$0.0560/call

Content Moderation

$0.002000/call

Sentiment Analysis

$0.002000/call

Categorization

$0.003100/call

Intent Detection

$0.001600/call

Copywriting

$0.0210/call

Blog Writing

$0.0820/call

Marketing Copy

$0.0316/call

Product Descriptions

$0.0160/call

Related pages

Frequently asked questions

How much does Claude Sonnet 5 cost for math reasoning per API call?

$0.0330 per call at 1,500 input / 3,000 output tokens.

Is Claude Sonnet 5 the best model for math reasoning?

Scores 90/100, rank #2 of 14 models. Top-ranked: Claude Fable 5 at 98/100.

What is the cheapest model for math reasoning with good quality?

Gemini 3.1 Pro scores 90/100 quality at $2.00/$12.00 per MTok. That is -18% less than Claude Sonnet 5 per call.

How can I reduce my Claude Sonnet 5 costs for math reasoning?

Three strategies: prompt caching (90% off input), batch API (50% off all tokens), prompt optimization. Combined: $0.0330 drops to about $0.001650 per call.

How does Claude Sonnet 5 pricing compare to other Claude models?

Claude also offers: Claude Opus 4.8 at $5.00/$25.00, Claude Sonnet 4.6 at $3.00/$15.00, Claude Haiku 4.5 at $1.00/$5.00, Claude Fable 5 at $10.00/$50.00.

What context window does Claude Sonnet 5 support?

200K tokens total. Typical math reasoning calls use 1,500 plus 3,000 tokens, well within limits.

Track what you spend on AI

This calculator estimates costs based on published rates only. Connect your API usage to see where your money goes and which models drive your spend.

Free spend teardownStart free dashboard