Compare · ModelsLive · 2 picked · head to head
Claude 3 Opus vs GPT-5.4 Nano
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5.4 Nano wins on 6/7 benchmarks
GPT-5.4 Nano wins 6 of 7 shared benchmarks. Leads in knowledge · general · math.
Category leads
knowledge·GPT-5.4 Nanogeneral·GPT-5.4 Nanomath·GPT-5.4 Nanocoding·GPT-5.4 Nano
Hype vs Reality
Attention vs performance
Claude 3 Opus
#261 by perf·no signal
GPT-5.4 Nano
#237 by perf·#4 by attention
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
7 benchmarks · 2 models
Claude 3 OpusGPT-5.4 Nano
Chess Puzzles
GPT-5.4 Nano leads by +26.3
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude 3 Opus
0.0
GPT-5.4 Nano
26.4
Dtbench
GPT-5.4 Nano leads by +31.1
Claude 3 Opus
36.0
GPT-5.4 Nano
67.1
GPQA diamond
GPT-5.4 Nano leads by +41.7
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude 3 Opus
29.6
GPT-5.4 Nano
71.3
Lmca
GPT-5.4 Nano leads by +23.4
Claude 3 Opus
20.0
GPT-5.4 Nano
43.4
OTIS Mock AIME 2024-2025
GPT-5.4 Nano leads by +83.1
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude 3 Opus
4.6
GPT-5.4 Nano
87.8
SimpleQA Verified
Claude 3 Opus leads by +0.9
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude 3 Opus
12.6
GPT-5.4 Nano
11.7
WeirdML
GPT-5.4 Nano leads by +30.0
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude 3 Opus
19.2
GPT-5.4 Nano
49.2
Full benchmark table
| Benchmark | Claude 3 Opus | GPT-5.4 Nano |
|---|---|---|
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 0.0 | 26.4 |
Dtbench | 36.0 | 67.1 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 29.6 | 71.3 |
Lmca | 20.0 | 43.4 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 4.6 | 87.8 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 12.6 | 11.7 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 19.2 | 49.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| $0.20 | $1.25 | 400K tokens (~200 books) | $4.63 |