Claude Opus 5.5vsClaude Sonnet 5.5
SI Score, benchmarks, price and context compared, with every number sourced.
83.7
SI Score
#1 of 124 ranked models
79.5
SI Score
#5 of 124 ranked models
Pillar by pillar
Claude Opus 5.5Claude Sonnet 5.5
89.0 Reasoning25% of score 86.4
93.3 Math25% of score 90.6
69.6 Coding25% of score 61.6
82.8 Preference25% of score 79.5
The basics
| Attribute | Claude Opus 5.5 | Claude Sonnet 5.5 |
|---|---|---|
| Input price, per 1M tokens | $4.00Anthropic models & pricingOfficial Claude API; lowest short-context on-demand tier; batch/cache/long-context rates excluded
Retrieved Oct 9, 2026 · factual citation Open source ↗ | $2.00Anthropic models & pricingOfficial Claude API; lowest short-context on-demand tier; batch/cache/long-context rates excluded
Retrieved Oct 9, 2026 · factual citation Open source ↗ |
| Output price, per 1M tokens | $20.00Anthropic models & pricingOfficial Claude API; lowest short-context on-demand tier; batch/cache/long-context rates excluded
Retrieved Oct 9, 2026 · factual citation Open source ↗ | $10.00Anthropic models & pricingOfficial Claude API; lowest short-context on-demand tier; batch/cache/long-context rates excluded
Retrieved Oct 9, 2026 · factual citation Open source ↗ |
| Context window | 1MAnthropic models & pricingOfficial Claude API; lowest short-context on-demand tier; batch/cache/long-context rates excluded
Retrieved Oct 9, 2026 · factual citation Open source ↗ | 1MAnthropic models & pricingOfficial Claude API; lowest short-context on-demand tier; batch/cache/long-context rates excluded
Retrieved Oct 9, 2026 · factual citation Open source ↗ |
| Released | Sep 22, 2026Anthropic models & pricingOfficial Claude API; lowest short-context on-demand tier; batch/cache/long-context rates excluded
Retrieved Oct 9, 2026 · factual citation Open source ↗ | Sep 28, 2026Anthropic models & pricingOfficial Claude API; lowest short-context on-demand tier; batch/cache/long-context rates excluded
Retrieved Oct 9, 2026 · factual citation Open source ↗ |
| Weights | Closed | Closed |
Highlighted values are the lower price or the larger context window.
Shared benchmarks
23 in common. Best result ahead: Claude Opus 5.5 on 12, Claude Sonnet 5.5 on 6Each side shows its best published result. Results can come from different settings or harnesses, so open a row to compare like with like before reading much into a small gap.
FrontierMath Tier 4 (v2) 95% 80.5%
Claude Opus 5.5
- max effort95%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 22, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗
Claude Sonnet 5.5
- max effort80.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 29, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗
FrontierMath Tiers 1–3 (v2) 91.2% 88.8%
Claude Opus 5.5
- max effort91.2%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 22, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗
Claude Sonnet 5.5
- max effort88.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 29, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗
GPQA Diamond 90.6% 95.6%
Claude Opus 5.5
- max effort90.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 22, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗
Claude Sonnet 5.5
- max effort95.6%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 29, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗
Humanity's Last Exam (with tools) 67.7% 64.5%
Claude Opus 5.5
- max effort67.7%Official model cards via models.devLab-reported; metric score; transcribed by MIT models.dev catalog; not independently evaluated [variant] max effort; with tools; production safeguards with fallbackPublished Sep 22, 2026
Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗
Claude Sonnet 5.5
- Published result64.5%Official model cards via models.devLab-reported; metric score; transcribed by MIT models.dev catalog; not independently evaluated [variant] with toolsPublished Sep 28, 2026
Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗
LiveBench Coding: code completion 87.0% 84.8%
Claude Opus 5.5
- Published result87.0%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result84.8%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Coding: code generation 91.5% 93.0%
Claude Opus 5.5
- Published result91.5%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result93.0%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Coding: JavaScript 72.7% 36.4%
Claude Opus 5.5
- Published result72.7%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result36.4%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Coding: Python 70% 35%
Claude Opus 5.5
- Published result70%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result35%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Coding: TypeScript 53.3% 46.7%
Claude Opus 5.5
- Published result53.3%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result46.7%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Math: AMPS Hard 99% 98%
Claude Opus 5.5
- Published result99%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result98%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Math: competition math 97.1% 97.1%
Claude Opus 5.5
- Published result97.1%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result97.1%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Math: integrals 99% 100%
Claude Opus 5.5
- Published result99%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result100%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Math: olympiad 92.2% 91.7%
Claude Opus 5.5
- Published result92.2%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result91.7%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Math: simplify 63% 72.4%
Claude Opus 5.5
- Published result63%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result72.4%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Reasoning: connections 99.3% 98.5%
Claude Opus 5.5
- Published result99.3%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result98.5%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Reasoning: consecutive events 90.4% 90.9%
Claude Opus 5.5
- Published result90.4%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result90.9%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Reasoning: logic with navigation 80% 80%
Claude Opus 5.5
- Published result80%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result80%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Reasoning: spatial 98% 98%
Claude Opus 5.5
- Published result98%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result98%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Reasoning: theory of mind 84.6% 69.2%
Claude Opus 5.5
- Published result84.6%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result69.2%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LiveBench Reasoning: zebra puzzles 100% 100%
Claude Opus 5.5
- Published result100%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
Claude Sonnet 5.5
- Published result100%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026
Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗
LMArena Text 1514.8 elo 1471.2 elo
Claude Opus 5.5
- Published result1514.8 eloLMArena / ArenaPublished source fact [variant] text / overallPublished Oct 8, 2026
Retrieved Oct 9, 2026 · CC-BY-4.0
Open source ↗
Claude Sonnet 5.5
- Published result1471.2 eloLMArena / ArenaPublished source fact [variant] text / overallPublished Oct 8, 2026
Retrieved Oct 9, 2026 · CC-BY-4.0
Open source ↗
OTIS Mock AIME 2024–2025 100% 100%
Claude Opus 5.5
- max effort100%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 22, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗
Claude Sonnet 5.5
- max effort100%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Sep 29, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗
Terminal-Bench 4.0 66.4% 70.6%
Claude Opus 5.5
- xhigh effort66.4%Official model cards via models.devLab-reported; metric score; transcribed by MIT models.dev catalog; not independently evaluated [variant] xhigh effort; production safeguards with fallback; 4.0Published Sep 22, 2026
Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗ - Setting 264.8%Terminal-BenchTerminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Claude Code; maxPublished Oct 6, 2026
Retrieved Oct 9, 2026 · Apache-2.0; factual citation
Open source ↗
Claude Sonnet 5.5
- Setting 170.6%Official model cards via models.devLab-reported; metric score; transcribed by MIT models.dev catalog; not independently evaluated [variant] 4.0Published Sep 28, 2026
Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗ - Setting 261.8%Terminal-BenchTerminal-Bench 4.0; published harness submission, 95% CI retained at source [variant] Claude Code; maxPublished Oct 6, 2026
Retrieved Oct 9, 2026 · Apache-2.0; factual citation
Open source ↗
Which should you choose?
- For coding, Claude Opus 5.5 leads by 8.0 points.
- For preference, Claude Opus 5.5 leads by 3.3 points.
- Claude Sonnet 5.5 costs less per input token ($2.00 vs $4.00 per 1M).
- Confidence is 100% for Claude Opus 5.5 and 93% for Claude Sonnet 5.5; sources still to report can move either score.
These follow from the numbers above. They're not a verdict on your use case.