Claude Opus 4.8 pricing for math reasoning

Quick start

Call Claude Opus 4.8 with the Anthropic Python SDK:

from anthropic import Anthropic

client = Anthropic()
response = client.messages.create(
    model="claude-opus-4-8",
    max_tokens=3000,
    messages=[{"role": "user", "content": "Your prompt here"}]
)
# Cost: ~$0.0825 per call at standard rates
# Input:  $5.00/MTok
# Output: $25.00/MTok

How Claude Opus 4.8 compares on cost

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

Cheapest viable option: GPT 5.6 Luna at $0.003900/call (95% less, 86/100 quality).

What drives math reasoning cost

Short problems, long solutions. Input is trivial and output is large because the derivation is the answer, so output rate is effectively the only price that matters. Models differ sharply in how much they show their working, which means two models at the same output rate can differ substantially in real cost per problem.

For Claude Opus 4.8 specifically, that means output tokens cost 5.0x more than input, at $25.00 vs $5.00 per MTok. At this use case’s 1,500/3,000 token split, output accounts for 91% of the $0.0825 per-call cost.

Claude Opus 4.8 quality benchmarks for math reasoning

Performance across all task types

Claude Opus 4.8 scores by task type. Math Reasoning highlighted.

Confidence scores are a third-party benchmark snapshot from week 26 of 2026. This snapshot is 17 weeks old, so treat the scores as directional and validate against your own evals.

About Claude Opus 4.8

Other models from Claude

You can compare Claude Opus 4.8 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.

Frequently asked questions