LLM Comparison

Claude Opus 4.8 vs Claude Sonnet 5: benchmark scores, pricing & comparison.

Side-by-side Claude Opus 4.8 vs Claude Sonnet 5 comparison across SWE-bench, GPQA, HLE, Terminal-Bench, coding agent scores, token pricing, context window, and AskClash RWT. Green marks the winner on each benchmark.

Verdict: Claude Opus 4.8 ranks higher (#25 vs #39) and wins 11 of 12 shared benchmarks; Claude Sonnet 5 costs less ($3.00/$15.0 vs $5.00/$25.0 per 1M tokens).

Rank #25 vs #39AskClash overall scores 54.8 vs 47.1.
Pricing $5.00/$25.0 vs $3.00/$15.0Input and output token prices per 1M tokens when published.
Proprietary vs ProprietaryAnthropic vs Anthropic.

Claude Opus 4.8 vs Claude Sonnet 5 benchmark comparison

Green cells highlight the winning model for each metric. Scores are cached from the AskClash LLM leaderboard snapshot.

MetricClaude Opus 4.8Claude Sonnet 5
Overall Score54.847.1
Leaderboard Rank#25#39
ACB60.160.8
AA Intelligence Index42.038.4
AutomationBench-AA46.0—
RWT9.08.5
HLE57.957.4
GPQA93.691.1
IFEval62.2—
SWE-bench88.685.2
SWE-Pro69.263.2
Terminal-Bench 4.021.714.1
DeepSWE59.053.8
GDPval-AA1889.81618.0
OSWorld83.481.2
MCP Atlas82.2—
CharXiv89.988.3
MMMU-Pro—77.3
ARC-AGI 272.1—
Tau294.4—
Input Price (per 1M tokens)$5.00$3.00
Output Price (per 1M tokens)$25.0$15.0
Context Window1M1M
Benchmarks Published1310

Claude Opus 4.8 vs Claude Sonnet 5 head-to-head charts

Claude Opus 4.8 leads 11 and Claude Sonnet 5 leads 1 of 12 shared benchmarks. Claude Sonnet 5 is cheaper on a defined 1M-input / 200K-output workload. Charts show only benchmarks both models publish.

Claude Opus 4.8Claude Sonnet 5
Overall
54.8Claude Opus 4.8
47.1Claude Sonnet 5
ACB
60.1Claude Opus 4.8
60.8Claude Sonnet 5
AA Intelligence Index
42.0Claude Opus 4.8
38.4Claude Sonnet 5
RWT
9.0Claude Opus 4.8
8.5Claude Sonnet 5
HLE
57.9Claude Opus 4.8
57.4Claude Sonnet 5
GPQA
93.6Claude Opus 4.8
91.1Claude Sonnet 5
SWE-bench
88.6Claude Opus 4.8
85.2Claude Sonnet 5
SWE-Pro
69.2Claude Opus 4.8
63.2Claude Sonnet 5
Terminal-Bench 4.0
21.7Claude Opus 4.8
14.1Claude Sonnet 5
DeepSWE
59.0Claude Opus 4.8
53.8Claude Sonnet 5
GDPval-AA
1889.8Claude Opus 4.8
1618.0Claude Sonnet 5
OSWorld
83.4Claude Opus 4.8
81.2Claude Sonnet 5
Claude Opus 4.8
Input$5.00
Output$25.0
Workload$10
Context1M
Claude Sonnet 5
Input$3.00
Output$15.0
Workload$6.00
Context1M

Workload = published cost of 1M input + 200K output tokens. Open the live leaderboard for interactive compare charts.

More Claude Opus 4.8 and Claude Sonnet 5 comparisons

Explore how Claude Opus 4.8 and Claude Sonnet 5 stack up against other top-ranked LLMs.

How to read this comparison

Benchmark scores

Higher is better for all benchmark scores (SWE-bench, GPQA, HLE, Terminal-Bench, etc.). Green marks the model with the higher score.

Token pricing

Lower is better for input and output prices. Green marks the cheaper model per 1M tokens.

Coverage matters

Models with fewer disclosed benchmark cells may have inflated percentile scores. Check the benchmark cell count for context.

This comparison page is generated from the AskClash LLM leaderboard cache. Open the live leaderboard for real-time scores and interactive filtering.