Claude Opus 4.7 vs Claude Opus 4.8

Benchmarks, API pricing and specs, head to head. Data updated 2026-06-10.

Claude Opus 4.7

Anthropic · Mar 2026

The early-2026 Opus release that pushed SWE-bench past 82% and extended the 1M context to the whole Opus line.

Claude Opus 4.8

Anthropic · May 2026

Anthropic's flagship Opus with adaptive thinking, effort controls and a 1M context — 88.6% on SWE-bench Verified at $5/$25.

The verdict

Claude Opus 4.8 wins 1 of the 1 benchmarks these models share, against 0 for Claude Opus 4.7.

Benchmark head-to-head 01

SWE-bench Verified
82%88.6%
Claude Opus 4.7Claude Opus 4.8

Specs & pricing

Claude Opus 4.7Claude Opus 4.8
modhub Index
Input price / 1M$5$5
Output price / 1M$25$25
Context window1M1M
Max output64K64K
Open weightsnono
Reasoning modelyesyes
Multimodal inputtext, imagetext, image
Knowledge cutoffNov 2025Mar 2026
ReleasedMar 2026May 2026
Example monthly cost*$87.50$87.50

* 10M input + 1.5M output tokens per month at list prices, no caching. Green = better value on that row.

Frequently asked questions

Which is better, Claude Opus 4.7 or Claude Opus 4.8?
Claude Opus 4.8 wins 1 of the 1 benchmarks these models share, against 0 for Claude Opus 4.7.
Which is cheaper, Claude Opus 4.7 or Claude Opus 4.8?
Claude Opus 4.7 costs $5/$25 per million input/output tokens, while Claude Opus 4.8 costs $5/$25. For a typical workload of 10M input and 1.5M output tokens per month, that's $87.50 versus $87.50.
Which model is better for coding, Claude Opus 4.7 or Claude Opus 4.8?
On SWE-bench Verified — the standard agentic-coding benchmark — Claude Opus 4.8 scores 88.6% versus 82% for Claude Opus 4.7, making Claude Opus 4.8 the stronger pick for coding agents.

More comparisons