Claude Opus 5 vs Claude Opus 4.8
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
by Anthropic
|
by Anthropic
|
|---|---|---|
| Overview | ||
| Status | Current flagship | Current flagship |
| Released | Jul 24, 2026 | May 28, 2026 |
| Pricing per million tokens | ||
| Input | $5.00/Mtok | $5.00/Mtok |
| Output | $25.00/Mtok | $25.00/Mtok |
| Blended avg | $15.00/Mtok | $15.00/Mtok |
| Cached input | $0.500/Mtok | $0.500/Mtok |
| Specifications | ||
| Context window | 1M tokens | 1M tokens |
| Parameters | Proprietary | Proprietary |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
textimage
|
textimage
|
| Benchmarks source: Vellum LLM Leaderboard | ||
| Avg benchmark score | 77.8 | 81.3 |
| Perf / dollar | 5.2 | 5.4 |
| Humanity's Last Exam | 64.7 | 57.9 |
| AutoBench | 26 | 15.5 |
| GPQA Diamond | — | 93.6 |
| SWE-Bench Verified | — | 88.6 |
| Terminal-Bench 2.1 | — | 74.6 |
| MCP Atlas | — | 82.2 |
| BrowseComp | 90.8 | — |
| OSWorld-Verified | — | 83.4 |
| HumanEval | — | 88.6 |
| Independent composite scores each on its own scale — not part of the average above | ||
| AA Intelligence Index | 60.7 | 55.7 |
| AA Agentic Index | 55.3 | 47.2 |
| AA-Omniscience Index (−100–100) | 31.3 | 27.4 |
| GDPval-AA v2 | 1861.5 ELO | 1593.1 ELO |
| LMSYS Chatbot Arena | — | 1474.4 ELO |
| Providers | ||
| Available from |
Anthropic — $5.00/$25.00/Mtok
|
Anthropic — $5.00/$25.00/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | Claude Opus 5 | Claude Opus 4.8 | Savings |
|---|---|---|---|
| 1M tokens | $15 | $15 | $0 (0%) |
| 10M tokens | $150 | $150 | $0 (0%) |
| 100M tokens | $1500 | $1500 | $0 (0%) |
| 1000M tokens | $15000 | $15000 | $0 (0%) |
The cheaper option here — Claude Opus 5 costs $15.00/Mtok blended on Anthropic.
Summary
Claude Opus 5 by Anthropic costs $5.00/Mtok input and $25.00/Mtok output, with a 1M-token context window. It supports text, image input.
Claude Opus 4.8 by Anthropic costs $5.00/Mtok input and $25.00/Mtok output, with a 1M-token context window. It supports text, image input.
On benchmarks, Claude Opus 4.8 scores higher (81.3 vs 77.8) on average. In terms of value, Claude Opus 4.8 has better performance per dollar (5.4 vs 5.2).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.