Compare · ModelsLive · 2 picked · head to head
Claude Opus 5 vs Claude Opus 5.5
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Opus 5.5 wins on 20/21 benchmarks
Claude Opus 5.5 wins 20 of 21 shared benchmarks. Leads in agentic · reasoning · knowledge.
Category leads
agentic·Claude Opus 5.5reasoning·Claude Opus 5.5knowledge·Claude Opus 5.5general·Claude Opus 5.5coding·Claude Opus 5.5math·Claude Opus 5.5
Head to head · from the data · data as of 2026-11-17
- Benchmarks · Claude Opus 5 leads on 1 of 21 shared benchmarks (GPQA diamond), Claude Opus 5.5 on 20 (APEX-Agents, ARC-AGI, ARC-AGI-2, Chess Puzzles, 16 more)
- Price · Claude Opus 5 input $5.00/M vs Claude Opus 5.5 $4.00/M, Claude Opus 5.5 is 20% cheaper (OpenRouter list price, as of 2026-11-17)
- Cheapest provider · Claude Opus 5 at Amazon Bedrock and 4 other providers $5.00/M input (of 11 offers) as of 2026-11-17 · Claude Opus 5.5 at Amazon Bedrock and 4 other providers $4.00/M input (of 11 offers) as of 2026-11-17
- Context · Claude Opus 5 and Claude Opus 5.5 both 1.0M tokens
- Released · Claude Opus 5 Jul 24, 2026 vs Claude Opus 5.5 Sep 22, 2026 · 2 months apart
- Gecko tests · Censorship Index: Claude Opus 5 refuses or dodges 2.6% vs Claude Opus 5.5 0% · Knowledge Horizon: Claude Opus 5 knows news up to Mar 2026 vs Claude Opus 5.5 May 2026
- Price history · Claude Opus 5 OpenRouter input list price had no move of 5% or more in the 90 days to 2026-11-17 · Claude Opus 5.5 OpenRouter input list price had no move of 5% or more in the 90 days to 2026-11-17
Sources · scores from public leaderboards, list prices and provider offers from OpenRouter, price moves from the BenchGecko changelog (5% or more, held in the next daily check), Gecko tests run by BenchGecko.
Hype vs Reality
Attention vs performance
Claude Opus 5
#25 by perf·#8 by attention
Claude Opus 5.5
#11 by perf·#8 by attention
Best value
Claude Opus 5.5
1.3x better value than Claude Opus 5
Claude Opus 5
4.6 pts/$
$15.00/M
Claude Opus 5.5
6.2 pts/$
$12.00/M
Vendor risk
Who is behind the model
Anthropic
$1.40T·Tier 1
Anthropic
$1.40T·Tier 1
Head to head
21 benchmarks · 2 models
Claude Opus 5Claude Opus 5.5
APEX-Agents
Claude Opus 5.5 leads by +7.7
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Opus 5
65.8
Claude Opus 5.5
73.5
ARC-AGI
Claude Opus 5.5 leads by +1.0
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Claude Opus 5
97.5
Claude Opus 5.5
98.5
ARC-AGI-2
Claude Opus 5.5 leads by +2.9
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Claude Opus 5
90.4
Claude Opus 5.5
93.3
Chess Puzzles
Claude Opus 5.5 leads by +25.3
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Opus 5
40.0
Claude Opus 5.5
65.3
Dtbench
Claude Opus 5.5 leads by +1.8
Claude Opus 5
96.5
Claude Opus 5.5
98.2
Ebr Bench
Claude Opus 5.5 leads by +25.7
Claude Opus 5
45.7
Claude Opus 5.5
71.4
Frontiercode
Claude Opus 5.5 leads by +1.3
Claude Opus 5
53.4
Claude Opus 5.5
54.6
FrontierMath-Tier-4-v2-Private
Claude Opus 5.5 leads by +21.8
Claude Opus 5
73.2
Claude Opus 5.5
95.0
FrontierMath-Tiers-1-3-v2-Private
Claude Opus 5.5 leads by +5.9
Claude Opus 5
90.1
Claude Opus 5.5
96.0
Frontierswe
Claude Opus 5.5 leads by +10.3
Claude Opus 5
52.0
Claude Opus 5.5
62.3
Furniture Assembly
Claude Opus 5.5 leads by +32.1
Claude Opus 5
44.0
Claude Opus 5.5
76.2
GPQA diamond
Claude Opus 5 leads by +4.6
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Opus 5
95.7
Claude Opus 5.5
91.1
Lmca
Claude Opus 5.5 leads by +4.5
Claude Opus 5
75.8
Claude Opus 5.5
80.3
Mirrorcode
Claude Opus 5.5 leads by +7.8
Claude Opus 5
87.8
Claude Opus 5.5
95.6
Mystery Game Puzzles
Claude Opus 5.5 leads by +13.2
Claude Opus 5
54.8
Claude Opus 5.5
68.0
OTIS Mock AIME 2024-2025
Claude Opus 5.5 leads by +1.1
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Opus 5
98.9
Claude Opus 5.5
100.0
PostTrainBench
Claude Opus 5.5 leads by +8.1
Claude Opus 5
35.7
Claude Opus 5.5
43.8
Proofbench
Claude Opus 5.5 leads by +1.0
Claude Opus 5
99.0
Claude Opus 5.5
100.0
SimpleBench
Claude Opus 5.5 leads by +9.4
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Opus 5
76.7
Claude Opus 5.5
86.1
SimpleQA Verified
Claude Opus 5.5 leads by +12.3
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Opus 5
59.9
Claude Opus 5.5
72.2
WeirdML v3
Claude Opus 5.5 leads by +19.2
Claude Opus 5
36.1
Claude Opus 5.5
55.4
Full benchmark table
| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 65.8 | 73.5 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 97.5 | 98.5 |
ARC-AGI-2 ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data. | 90.4 | 93.3 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 40.0 | 65.3 |
Dtbench | 96.5 | 98.2 |
Ebr Bench | 45.7 | 71.4 |
Frontiercode | 53.4 | 54.6 |
FrontierMath-Tier-4-v2-Private | 73.2 | 95.0 |
FrontierMath-Tiers-1-3-v2-Private | 90.1 | 96.0 |
Frontierswe | 52.0 | 62.3 |
Furniture Assembly | 44.0 | 76.2 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 95.7 | 91.1 |
Lmca | 75.8 | 80.3 |
Mirrorcode | 87.8 | 95.6 |
Mystery Game Puzzles | 54.8 | 68.0 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 98.9 | 100.0 |
PostTrainBench | 35.7 | 43.8 |
Proofbench | 99.0 | 100.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 76.7 | 86.1 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 59.9 | 72.2 |
WeirdML v3 | 36.1 | 55.4 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $5.00 | $25.00 | 1.0M tokens (~500 books) | $100.00 | |
| $4.00 | $20.00 | 1.0M tokens (~500 books) | $80.00 |
Claude Opus 5 vs Claude Opus 5.5 · questions answered from the data
Claude Opus 5 leads on 1 of 21 shared benchmarks (GPQA diamond), Claude Opus 5.5 on 20 (APEX-Agents, ARC-AGI, ARC-AGI-2, Chess Puzzles, 16 more)
People also compared