Compare · ModelsLive · 2 picked · head to head
Claude Opus 4.8 vs GPT-5.5
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5.5 wins on 21/29 benchmarks
GPT-5.5 wins 21 of 29 shared benchmarks. Leads in agentic · reasoning · arena.
Category leads
agentic·GPT-5.5reasoning·GPT-5.5arena·GPT-5.5knowledge·GPT-5.5coding·Claude Opus 4.8general·GPT-5.5math·GPT-5.5
Head to head · from the data · data as of 2026-11-17
- Benchmarks · Claude Opus 4.8 leads on 8 of 29 shared benchmarks (Frontiercode, GSO-Bench, Lmca, Mirrorcode, 4 more), GPT-5.5 on 21 (APEX-Agents, ARC-AGI, ARC-AGI-2, Chatbot Arena Elo · Overall, 17 more)
- Price · Claude Opus 4.8 input $5.00/M vs GPT-5.5 $5.00/M, the same input price (OpenRouter list price, as of 2026-11-17)
- Cheapest provider · Claude Opus 4.8 at Amazon Bedrock and 4 other providers $5.00/M input (of 11 offers) as of 2026-11-17 · GPT-5.5 at OpenAI $2.50/M input (of 7 offers) as of 2026-11-17
- Context · Claude Opus 4.8 1.0M tokens vs GPT-5.5 1.1M tokens
- Released · Claude Opus 4.8 May 27, 2026 vs GPT-5.5 Apr 24, 2026 · 1 month apart
- Price history · Claude Opus 4.8 OpenRouter input list price had no move of 5% or more in the 90 days to 2026-11-17 · GPT-5.5 OpenRouter input list price had no move of 5% or more in the 90 days to 2026-11-17
Sources · scores from public leaderboards, list prices and provider offers from OpenRouter, price moves from the BenchGecko changelog (5% or more, held in the next daily check), Gecko tests run by BenchGecko.
Hype vs Reality
Attention vs performance
Claude Opus 4.8
#72 by perf·#8 by attention
GPT-5.5
#64 by perf·#11 by attention
Best value
Claude Opus 4.8
1.2x better value than GPT-5.5
Claude Opus 4.8
3.8 pts/$
$15.00/M
GPT-5.5
3.3 pts/$
$17.50/M
Vendor risk
Who is behind the model
Anthropic
$1.40T·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
29 benchmarks · 2 models
Claude Opus 4.8GPT-5.5
APEX-Agents
GPT-5.5 leads by +6.2
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Opus 4.8
48.9
GPT-5.5
55.1
ARC-AGI
GPT-5.5 leads by +2.5
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Claude Opus 4.8
92.5
GPT-5.5
95.0
ARC-AGI-2
GPT-5.5 leads by +12.9
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 4.8
72.1
GPT-5.5
85.0
Chatbot Arena Elo · Overall
GPT-5.5 leads by +2.2
Claude Opus 4.8
1474.5
GPT-5.5
1476.8
Chess Puzzles
GPT-5.5 leads by +11.6
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Opus 4.8
49.5
GPT-5.5
61.1
DeepResearch Bench
GPT-5.5 leads by +3.8
DeepResearch Bench · evaluates AI on complex multi-step research tasks requiring information gathering, synthesis, and producing comprehensive analyses.
Claude Opus 4.8
50.2
GPT-5.5
54.0
Deepswe
GPT-5.5 leads by +8.1
Claude Opus 4.8
59.0
GPT-5.5
67.0
Dtbench
GPT-5.5 leads by +1.8
Claude Opus 4.8
91.5
GPT-5.5
93.3
Ebr Bench
GPT-5.5 leads by +5.7
Claude Opus 4.8
28.6
GPT-5.5
34.3
Frontiercode
Claude Opus 4.8 leads by +3.5
Claude Opus 4.8
46.5
GPT-5.5
43.0
FrontierMath-2025-02-28-Private
GPT-5.5 leads by +4.5
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
Claude Opus 4.8
47.2
GPT-5.5
51.7
FrontierMath-Tier-4-2025-07-01-Private
GPT-5.5 leads by +4.1
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
Claude Opus 4.8
31.3
GPT-5.5
35.4
FrontierMath-Tier-4-v2-Private
GPT-5.5 leads by +16.4
Claude Opus 4.8
56.1
GPT-5.5
72.5
FrontierMath-Tiers-1-3-v2-Private
GPT-5.5 leads by +5.5
Claude Opus 4.8
84.2
GPT-5.5
89.8
Furniture Assembly
GPT-5.5 leads by +2.4
Claude Opus 4.8
17.9
GPT-5.5
20.2
GPQA diamond
GPT-5.5 leads by +4.1
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Opus 4.8
91.7
GPT-5.5
95.8
GSO-Bench
Claude Opus 4.8 leads by +6.9
GSO-Bench · evaluates AI models on real-world open-source software engineering tasks, testing the ability to understand and resolve actual GitHub issues.
Claude Opus 4.8
47.1
GPT-5.5
40.2
Lmca
Claude Opus 4.8 leads by +3.8
Claude Opus 4.8
67.7
GPT-5.5
63.9
Mirrorcode
Claude Opus 4.8 leads by +42.2
Claude Opus 4.8
65.6
GPT-5.5
23.3
Mystery Game Puzzles
GPT-5.5 leads by +22.0
Claude Opus 4.8
29.5
GPT-5.5
51.5
Osworld 2 0
Claude Opus 4.8 leads by +7.6
Claude Opus 4.8
20.6
GPT-5.5
13.0
OTIS Mock AIME 2024-2025
GPT-5.5 leads by +1.7
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Opus 4.8
98.3
GPT-5.5
100.0
PostTrainBench
Claude Opus 4.8 leads by +7.1
Claude Opus 4.8
31.0
GPT-5.5
23.9
Proofbench
Claude Opus 4.8 leads by +19.0
Claude Opus 4.8
69.0
GPT-5.5
50.0
Remote Labor Index
Claude Opus 4.8 leads by +2.1
Claude Opus 4.8
8.3
GPT-5.5
6.3
SimpleBench
GPT-5.5 leads by +5.0
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Opus 4.8
57.8
GPT-5.5
62.8
SimpleQA Verified
GPT-5.5 leads by +10.0
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Opus 4.8
53.0
GPT-5.5
63.0
Surface Evolver Bench
GPT-5.5 leads by +0.6
Claude Opus 4.8
87.5
GPT-5.5
88.1
WeirdML
GPT-5.5 leads by +2.0
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Opus 4.8
82.9
GPT-5.5
84.9
Full benchmark table
| Benchmark | Claude Opus 4.8 | GPT-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. | 48.9 | 55.1 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 92.5 | 95.0 |
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. | 72.1 | 85.0 |
Chatbot Arena Elo · Overall | 1474.5 | 1476.8 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 49.5 | 61.1 |
DeepResearch Bench DeepResearch Bench · evaluates AI on complex multi-step research tasks requiring information gathering, synthesis, and producing comprehensive analyses. | 50.2 | 54.0 |
Deepswe | 59.0 | 67.0 |
Dtbench | 91.5 | 93.3 |
Ebr Bench | 28.6 | 34.3 |
Frontiercode | 46.5 | 43.0 |
FrontierMath-2025-02-28-Private FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning. | 47.2 | 51.7 |
FrontierMath-Tier-4-2025-07-01-Private FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning. | 31.3 | 35.4 |
FrontierMath-Tier-4-v2-Private | 56.1 | 72.5 |
FrontierMath-Tiers-1-3-v2-Private | 84.2 | 89.8 |
Furniture Assembly | 17.9 | 20.2 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 91.7 | 95.8 |
GSO-Bench GSO-Bench · evaluates AI models on real-world open-source software engineering tasks, testing the ability to understand and resolve actual GitHub issues. | 47.1 | 40.2 |
Lmca | 67.7 | 63.9 |
Mirrorcode | 65.6 | 23.3 |
Mystery Game Puzzles | 29.5 | 51.5 |
Osworld 2 0 | 20.6 | 13.0 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 98.3 | 100.0 |
PostTrainBench | 31.0 | 23.9 |
Proofbench | 69.0 | 50.0 |
Remote Labor Index | 8.3 | 6.3 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 57.8 | 62.8 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 53.0 | 63.0 |
Surface Evolver Bench | 87.5 | 88.1 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 82.9 | 84.9 |
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 | |
| $5.00 | $30.00 | 1.1M tokens (~525 books) | $112.50 |
Claude Opus 4.8 vs GPT-5.5 · questions answered from the data
Claude Opus 4.8 leads on 8 of 29 shared benchmarks (Frontiercode, GSO-Bench, Lmca, Mirrorcode, 4 more), GPT-5.5 on 21 (APEX-Agents, ARC-AGI, ARC-AGI-2, Chatbot Arena Elo · Overall, 17 more)