NEWMEN
← back to all models

Anthropic: Claude Sonnet 4

anthropic/claude-sonnet-4

Anthropicclosed-weight1M context6 providersIntelligence 75.8

Cheapest provider

$3.00 / 1M

Amazon Bedrock

Fastest provider (p95)

No throughput data yet — populated as traffic accumulates

Intelligence (composite)

75.8

MMLU-Pro · HumanEval · math · GPQA

Per-provider performance

Latency / throughput / uptime / price measured across providers over the last 30 minutes of live traffic. This is what proves “sourced cheapest” — Atlas mode draws on these per call to serve the cheapest path that holds quality.

ProviderQuantInput $/1MOutput $/1MLatency p50 / p95Throughput p50 / p95Uptime 30mSuccess
Amazon Bedrockfull· unknown$3.0000$15.0000
Amazon Bedrockfull· unknown$3.0000$15.0000100.00%
Anthropicfull· unknown$3.0000$15.0000
Googlefull· unknown$3.0000$15.0000100.00%
Googlefull· unknown$3.0000$15.0000
Googlefull· unknown$3.0000$15.0000

“—” means live telemetry hasn’t accumulated enough recent traffic for that endpoint. “undisclosed” means the provider serves the model but doesn’t expose the quantization label (typically running fp8 / int8 internally).

Intelligence breakdown

Composite score is a weighted average of public benchmarks (30% MMLU-Pro, 25% code pass@1, 25% math, 20% GPQA). Numbers come from model cards and the Artificial Analysis intelligence harness; missing components are renormalised over what’s present.

MMLU-Pro

84.0

broad reasoning

Code

66.0

pass@1 (HumanEval / LiveCodeBench)

AIME 2025

74.0

math accuracy

GPQA Diamond

78.0

hard reasoning

Source: Artificial Analysis (AA-aligned) — Claude Sonnet 4 (MMLU-Pro 84, GPQA-D 78, LiveCodeBench 66, AIME25 74); github.com/MiniMax-AI/MiniMax-M2

How Atlas mode sources Anthropic: Claude Sonnet 4

  • Strict mode — pin Anthropic: Claude Sonnet 4 exactly and we pass it straight through, sourced from the cheapest provider above. The same model, no substitutions — currently Amazon Bedrock at $3.00/1M.
  • Atlas mode — the default. Each call is auto-optimized for the cheapest path that holds quality, at least 5% off going direct from call one and climbing as it ramps. You always see which model served the call and exactly what you saved — thumbs-down anything you don’t like for a full refund.