Skip to content
SF·SIX
SF·SIX / MODEL COMPARISON

Claude Sonnet 5.5 vs Claude Opus 5.5

A side-by-side of live API prices, providers, speed, usage and benchmarks. No verdict: the right choice depends on your workload.

By the SF·SIX research desk, ColdAI · Updated · Prices checked · Methodology

Claude Sonnet 5.5 costs $2.00 per 1M input and $10.00 per 1M output tokens at the standard tier, via Anthropic; Claude Opus 5.5 costs $4.00 and $20.00, via Anthropic. At a 3:1 input:output mix, Claude Sonnet 5.5 is 50% cheaper than Claude Opus 5.5.

Why these two are compared

Both are Anthropic models released in September 2026: Sonnet 5.5 for well-scoped everyday work, Opus 5.5 as the flagship for demanding reasoning and long-horizon agentic work. The choice is usually a cost-per-task trade-off within one vendor.

Side by side

Side by side
MeasureClaude Sonnet 5.5Claude Opus 5.5
DeveloperAnthropicAnthropic
Released2026-09-282026-09-22
Standard price, input / output per 1M$2.00 / $10.00$4.00 / $20.00
Blended price (3:1)$4.00$8.00
Cheapest standard providerAnthropicAnthropic
Cache read per 1M$0.10$0.20
Providers at the standard tier55
Context window (tokens)1,000,0001,000,000
Max output (tokens)128,000128,000
Input modalitiestext, image, filetext, image, file
Long-prompt pricingNo threshold listedNo threshold listed
Fastest endpoint135 tok/sGoogle (US region)176 tok/sAnthropic (Fast tier)
Lowest latency942 msGoogle (Europe region)794 msGoogle (Europe region)
OpenRouter tokens (2026-10-16)396.5B933.37B
AA intelligence56.057.6
AA codingNot reportedNot reported
AA agenticNot reportedNot reported

“Not reported” means the source publishes no value. AA scores are the Artificial Analysis indexes reported through OpenRouter (CC BY 4.0), not SF·SIX measurements.

Cost for three example workloads

Cost for three example workloads
WorkloadClaude Sonnet 5.5 / monthClaude Opus 5.5 / month
Support chatbot50,000 requests × 1,500 in / 400 out tokens$350.00$700.00
RAG / research assistant20,000 requests × 8,000 in / 700 out tokens$460.00$920.00
Batch summarisation10,000 requests × 10,000 in / 500 out tokens$250.00$500.00

Standard headline prices; caching and long-context surcharges not included. Price your own workload.

Claude Sonnet 5.5 providers

Claude Sonnet 5.5 providers
Provider and termsInput / 1MOutput / 1MThroughputLatency
Amazon Bedrock$2.00$10.00106 tok/s2.40 s
Anthropic$2.00$10.0098 tok/s1.95 s
AzureGlobal endpoint$2.00$10.0099 tok/s1.05 s
Claude Platform on AWS$2.00$10.00101 tok/s1.95 s
GoogleGlobal endpoint$2.00$10.00112 tok/s3.37 s
AzureUS region$2.20$11.00Not reportedNot reported
GoogleEurope region$2.20$11.00100 tok/s942 ms
GoogleUS region$2.20$11.00135 tok/s1.10 s

All Claude Sonnet 5.5 prices and providers

Claude Opus 5.5 providers

Claude Opus 5.5 providers
Provider and termsInput / 1MOutput / 1MThroughputLatency
Amazon Bedrock$4.00$20.0075 tok/s4.61 s
Anthropic$4.00$20.0074 tok/s3.48 s
AzureGlobal endpoint$4.00$20.0078 tok/s2.84 s
Claude Platform on AWS$4.00$20.0076 tok/s4.23 s
GoogleGlobal endpoint$4.00$20.0080 tok/s2.95 s
Amazon Bedrockeu-west-1 region$4.40$22.0075 tok/s4.46 s
Amazon Bedrockus-east-1 region$4.40$22.0067 tok/s3.88 s
AzureUS region$4.40$22.00Not reportedNot reported
GoogleEurope region$4.40$22.0095 tok/s794 ms
GoogleUS region$4.40$22.0091 tok/s2.40 s

All Claude Opus 5.5 prices and providers

Related comparisons

Sources and method

Prices are USD per 1M tokens at the standard service tier in the default region: the cheapest qualifying provider per model, with flex, batch, priority and fast endpoints and unusually low outliers never used as the headline. “Blended” weights input and output 3:1. Prices: OpenRouter, observed at the times on each model page. Speed: OpenRouter, previous 30 minutes. Usage: OpenRouter rankings, latest full UTC day (CC BY 4.0).

Related pages