Compare · ModelsLive · 3 picked · head to head
GPT-5.4 Pro vs GPT-5.4 vs Claude Opus 4.8
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5.4 Pro wins on 9/17 benchmarks
GPT-5.4 Pro wins 9 of 17 shared benchmarks. Leads in reasoning · knowledge.
Category leads
reasoning·GPT-5.4 Proknowledge·GPT-5.4 Promath·Claude Opus 4.8coding·Claude Opus 4.8agentic·Claude Opus 4.8arena·Claude Opus 4.8
Hype vs Reality
Attention vs performance
GPT-5.4 Pro
#41 by perf·no signal
GPT-5.4
#61 by perf·no signal
Claude Opus 4.8
#47 by perf·no signal
Best value
GPT-5.4
1.6x better value than Claude Opus 4.8
GPT-5.4 Pro
0.6 pts/$
$105.00/M
GPT-5.4
6.7 pts/$
$8.75/M
Claude Opus 4.8
4.1 pts/$
$15.00/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
OpenAI
$840.0B·Tier 1
Anthropic
$380.0B·Tier 1
Head to head
17 benchmarks · 3 models
GPT-5.4 ProGPT-5.4Claude Opus 4.8
ARC-AGI
GPT-5.4 Pro leads by +0.8
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
GPT-5.4 Pro
94.5
GPT-5.4
93.7
Claude Opus 4.8
92.5
ARC-AGI-2
GPT-5.4 Pro leads by +9.4
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
GPT-5.4 Pro
83.3
GPT-5.4
74.0
Claude Opus 4.8
72.1
Chess Puzzles
GPT-5.4 Pro leads by +14.6
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GPT-5.4 Pro
58.6
GPT-5.4
44.0
Claude Opus 4.8
34.0
FrontierMath-2025-02-28-Private
GPT-5.4 Pro leads by +2.4
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GPT-5.4 Pro
50.0
GPT-5.4
47.6
Claude Opus 4.8
47.2
FrontierMath-Tier-4-2025-07-01-Private
GPT-5.4 Pro leads by +6.3
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
GPT-5.4 Pro
37.5
GPT-5.4
27.1
Claude Opus 4.8
31.3
GPQA diamond
GPT-5.4 Pro leads by +1.7
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-5.4 Pro
92.8
GPT-5.4
91.1
Claude Opus 4.8
88.0
SimpleQA Verified
GPT-5.4 Pro leads by +3.0
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GPT-5.4 Pro
47.8
GPT-5.4
44.8
Claude Opus 4.8
39.5
WeirdML
Claude Opus 4.8 leads by +5.2
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
GPT-5.4 Pro
57.4
GPT-5.4
77.7
Claude Opus 4.8
82.9
APEX-Agents
Claude Opus 4.8 leads by +6.6
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
GPT-5.4
35.9
Claude Opus 4.8
42.5
Chatbot Arena Elo · Coding
Claude Opus 4.8 leads by +131.1
GPT-5.4
1411.1
Claude Opus 4.8
1542.3
Chatbot Arena Elo · Overall
Claude Opus 4.8 leads by +10.1
GPT-5.4
1467.7
Claude Opus 4.8
1477.8
FrontierMath-Tier-4-v2-Private
Claude Opus 4.8 leads by +7.1
GPT-5.4
49.0
Claude Opus 4.8
56.1
FrontierMath-Tiers-1-3-v2-Private
Claude Opus 4.8 leads by +1.4
GPT-5.4
78.6
Claude Opus 4.8
80.0
HLE
GPT-5.4 Pro leads by +8.5
HLE (Humanity's Last Exam) · a reasoning benchmark designed to be the hardest public evaluation of AI. Questions span mathematics, physics, philosophy, and logic · curated to be at or beyond the frontier of human expert capability. Tested with and without tool augmentation. Claude Opus 4.7 scores 46.9% without tools and 54.7% with tools · making it one of the few benchmarks where the top score is below 60%.
GPT-5.4 Pro
41.5
GPT-5.4
33.0
OTIS Mock AIME 2024-2025
Claude Opus 4.8 leads by +3.0
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-5.4
95.3
Claude Opus 4.8
98.3
PostTrainBench
Claude Opus 4.8 leads by +17.0
GPT-5.4
20.2
Claude Opus 4.8
37.2
SimpleBench
GPT-5.4 Pro leads by +11.2
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
GPT-5.4 Pro
68.9
Claude Opus 4.8
57.8
Full benchmark table
| Benchmark | GPT-5.4 Pro | GPT-5.4 | Claude Opus 4.8 |
|---|---|---|---|
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 94.5 | 93.7 | 92.5 |
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. | 83.3 | 74.0 | 72.1 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 58.6 | 44.0 | 34.0 |
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. | 50.0 | 47.6 | 47.2 |
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. | 37.5 | 27.1 | 31.3 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 92.8 | 91.1 | 88.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 47.8 | 44.8 | 39.5 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 57.4 | 77.7 | 82.9 |
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | — | 35.9 | 42.5 |
Chatbot Arena Elo · Coding | — | 1411.1 | 1542.3 |
Chatbot Arena Elo · Overall | — | 1467.7 | 1477.8 |
FrontierMath-Tier-4-v2-Private | — | 49.0 | 56.1 |
FrontierMath-Tiers-1-3-v2-Private | — | 78.6 | 80.0 |
HLE HLE (Humanity's Last Exam) · a reasoning benchmark designed to be the hardest public evaluation of AI. Questions span mathematics, physics, philosophy, and logic · curated to be at or beyond the frontier of human expert capability. Tested with and without tool augmentation. Claude Opus 4.7 scores 46.9% without tools and 54.7% with tools · making it one of the few benchmarks where the top score is below 60%. | 41.5 | 33.0 | — |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | — | 95.3 | 98.3 |
PostTrainBench | — | 20.2 | 37.2 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 68.9 | — | 57.8 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $30.00 | $180.00 | 1.1M tokens (~525 books) | $675.00 | |
| $2.50 | $15.00 | 1.1M tokens (~525 books) | $56.25 | |
| $5.00 | $25.00 | 1.0M tokens (~500 books) | $100.00 |