Anthropic Claude Sonnet 4.6

Anthropic Claude Sonnet 4.6: $3/1M input tokens, 280 tok/s throughput, 700ms latency. Compare on BenchNode.io.

per 1M input tokens

$15 / 1M output

Specifications

model id: claude-sonnet-4.6
model family: Claude 4
parameters: undisclosed
context window tokens: 200,000
modality: text + vision
reasoning: No

Performance

Latency (TTFT) 700 ms

lower is better

Uptime SLA 99.99%

Throughput 280 tok/s

700

ms latency

99.99%

uptime

Pricing Detail

input per 1m tokens usd: 3
output per 1m tokens usd: 15
free tier available: No
rate limit tpm: 160,000
rate limit rpm: 1,000
batch discount pct: 50
context caching input usd: 0.3

AI Analysis · gpt-4o-mini

Technical Verdict

The input cost of $3.0 / 1M tokens and output cost of $15.0 / 1M tokens indicate a premium pricing tier for this model class, while the throughput of 280 tok/s is relatively fast, benefiting bulk async jobs with a 50% discount on batch pricing and context caching at $0.3/1M cached input tokens. The rate limits of 160,000 TPM and 1000 RPM suggest suitability for high-volume production, while the first-token latency of 700 ms leans towards batch use rather than real-time applications.

Ideal Use Case

This API tier is suitable for a batch document processing pipeline with a team of >10 engineers requiring >500k daily requests and tolerating moderate latency.