Compare · ModelsLive · 2 picked · head to head
GPT-5.4 vs Gemini 3 Pro
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5.4 wins on 13/16 benchmarks
GPT-5.4 wins 13 of 16 shared benchmarks. Leads in speed · agentic · reasoning.
Category leads
speed·GPT-5.4agentic·GPT-5.4reasoning·GPT-5.4arena·Gemini 3 Proknowledge·GPT-5.4math·GPT-5.4coding·GPT-5.4
Hype vs Reality
Attention vs performance
GPT-5.4
#44 by perf·no signal
Gemini 3 Pro
#38 by perf·no signal
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Google DeepMind
$4.00T·Tier 1
Head to head
16 benchmarks · 2 models
GPT-5.4Gemini 3 Pro
Artificial Analysis · Agentic Index
GPT-5.4 leads by +24.4
GPT-5.4
69.4
Gemini 3 Pro
45.0
Artificial Analysis · Coding Index
GPT-5.4 leads by +17.9
GPT-5.4
57.3
Gemini 3 Pro
39.4
Artificial Analysis · Quality Index
GPT-5.4 leads by +15.9
GPT-5.4
57.2
Gemini 3 Pro
41.3
APEX-Agents
GPT-5.4 leads by +17.5
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
Gemini 3 Pro
18.4
ARC-AGI
GPT-5.4 leads by +18.7
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
GPT-5.4
93.7
Gemini 3 Pro
75.0
ARC-AGI-2
GPT-5.4 leads by +42.8
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
74.0
Gemini 3 Pro
31.1
Chatbot Arena Elo · Overall
Gemini 3 Pro leads by +20.4
GPT-5.4
1465.8
Gemini 3 Pro
1486.2
Chess Puzzles
GPT-5.4 leads by +13.0
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GPT-5.4
44.0
Gemini 3 Pro
31.0
FrontierMath-2025-02-28-Private
GPT-5.4 leads by +10.0
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GPT-5.4
47.6
Gemini 3 Pro
37.6
FrontierMath-Tier-4-2025-07-01-Private
GPT-5.4 leads by +8.4
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
27.1
Gemini 3 Pro
18.8
GPQA diamond
GPT-5.4 leads by +0.9
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
91.1
Gemini 3 Pro
90.2
OTIS Mock AIME 2024-2025
GPT-5.4 leads by +3.9
OTIS Mock AIME 2024–2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-5.4
95.3
Gemini 3 Pro
91.4
PostTrainBench
GPT-5.4 leads by +2.1
GPT-5.4
20.2
Gemini 3 Pro
18.1
SimpleQA Verified
Gemini 3 Pro leads by +28.1
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GPT-5.4
44.8
Gemini 3 Pro
72.9
SWE-Bench verified
GPT-5.4 leads by +3.9
GPT-5.4
76.9
Gemini 3 Pro
72.9
WeirdML
Gemini 3 Pro leads by +12.5
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
GPT-5.4
57.4
Gemini 3 Pro
69.9
Full benchmark table
| Benchmark | GPT-5.4 | Gemini 3 Pro |
|---|---|---|
Artificial Analysis · Agentic Index | 69.4 | 45.0 |
Artificial Analysis · Coding Index | 57.3 | 39.4 |
Artificial Analysis · Quality Index | 57.2 | 41.3 |
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 | 18.4 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 93.7 | 75.0 |
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. | 74.0 | 31.1 |
Chatbot Arena Elo · Overall | 1465.8 | 1486.2 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 44.0 | 31.0 |
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. | 47.6 | 37.6 |
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. | 27.1 | 18.8 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 91.1 | 90.2 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024–2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 95.3 | 91.4 |
PostTrainBench | 20.2 | 18.1 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 44.8 | 72.9 |
SWE-Bench verified | 76.9 | 72.9 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 57.4 | 69.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| GPT-5.4 | $2.50 | $15.00 | 1.1M tokens (~525 books) | $56.25 |
| Gemini 3 Pro | — | — | — | — |