Claude Opus 4.5 vs Claude Opus 4.8: Benchmarks, Pricing & Capabilities Compared
TL;DR — Claude Opus 4.5 wins for general use · Claude Opus 4.8 wins for reasoning + cost + long-context.
Claude Opus 4.5 Anthropic
- Released
- 2025-11-24
- Context window
- 500K tokens
- Input price
- $15.00 / Mtok
- Output price
- $75.00 / Mtok
Key features
- First model to break 80.9% on SWE-Bench Verified
- 67% price reduction vs previous Opus
- Extended reasoning capabilities
Claude Opus 4.8 Anthropic
- Released
- 2026-05-28
- Context window
- 1M tokens
- Input price
- $6.25 / Mtok
- Output price
- $25.00 / Mtok
Benchmark comparison
| Benchmark | Claude Opus 4.5 | Claude Opus 4.8 |
|---|---|---|
| GPQA Diamond | 86.6% | 92.0% ✓ |
| HLE | 28.4% | 45.7% ✓ |
| IF-Bench | 58.0% | 62.2% ✓ |
| LiveCodeBench Reasoning | 74.0% ✓ | 67.7% |
| SciCode | 49.5% | 53.5% ✓ |
| TAU2-bench | 89.5% | 94.4% ✓ |
| TerminalBench-Hard | 47.0% | 58.3% ✓ |
Pricing comparison
| Metric | Claude Opus 4.5 | Claude Opus 4.8 |
|---|---|---|
| Input ($/Mtok) | $15.00 | $6.25 |
| Output ($/Mtok) | $75.00 | $25.00 |
| Cached input ($/Mtok) | $1.50 | — |
| Cost per 1M-token roundtrip (1M in + 1M out) | $90.00 | $31.25 |
Context window & modalities
| Attribute | Claude Opus 4.5 | Claude Opus 4.8 |
|---|---|---|
| Context window | 500K tokens | 1M tokens |
| Input modalities | text, image, PDF | text, image |
| Output modalities | text | text |
| Knowledge cutoff | 2025-08 | — |
Verdict by use case
Coding
Insufficient data
Basis: SWE-bench
No shared coding benchmark.
Reasoning
→ Claude Opus 4.8
Basis: GPQA Diamond
Claude Opus 4.5 86.6% vs Claude Opus 4.8 92% on GPQA Diamond.
Math
Insufficient data
Basis: MATH / AIME
No shared math benchmark.
Long context
→ Claude Opus 4.8
Basis: Context window
Claude Opus 4.5 500K tokens vs Claude Opus 4.8 1M tokens.
Cost
→ Claude Opus 4.8
Basis: Input $/Mtok
Claude Opus 4.5 $15/Mtok vs Claude Opus 4.8 $6.25/Mtok input.
Changelog & releases
Claude Opus 4.8
Released 2026-05-28