Polarison

Anthropic

Claude Sonnet 5 pricing and benchmarks

Anthropic’s best combination of speed and intelligence.

Input
$2.00
per 1M tokens
Cached input
$0.20
per 1M tokens
Output
$10.00
per 1M tokens
Capability index
156.2
#9 of 16 tracked

How capable is Claude Sonnet 5?

Claude Sonnet 5 scores 156.2 on the Epoch Capabilities Index (confidence interval 153.6158.6), ranking #9 of the 16 rated models we track.

BenchmarkTestsScore
GPQA DiamondScience reasoning
87.4%
FrontierMath (Tiers 1–3)Advanced math
65.6%
SimpleQA VerifiedFactual accuracy
33.7%
ARC-AGI-2Abstract reasoning
DeepSWECoding
53.8%
APEX-AgentsAgentic work
32.5%

Source: Epoch AI (as “Claude Sonnet 5”), best recorded result, CC BY 4.0. A dash means no published score. Compare all models

What Claude Sonnet 5 costs in practice

WorkloadMonthly cost
Support chatbot
100,000 replies a month, ~1,500 input tokens (system prompt + history) and ~400 output tokens each.
$700.00
Document Q&A (RAG)
50,000 questions a month with ~8,000 tokens of retrieved context and ~600 output tokens each.
$1,100
Coding agent
10,000 agent steps a month, ~40,000 input tokens each (70% read from the prompt cache) and ~2,000 output tokens.
$496.00

Pricing details

  • The $2 / $10 launch price is now the standard price; the planned rise to $3 / $15 will not happen.
  • Cache writes cost $2.50 (5-minute cache) or $4 (1-hour cache) per 1M tokens.
  • Uses Anthropic’s newer tokenizer, which produces roughly 30% more tokens for the same text.

Estimate your Claude Sonnet 5 bill

Standard-tier list prices. Long-context rates apply automatically where the provider publishes a threshold. Excludes cache-write surcharges, taxes, batch and volume discounts.