Compare · ModelsLive · 2 picked · head to head
Claude 3.5 Sonnet vs GPT-5 Nano
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-5 Nano wins on 11/11 benchmarks
GPT-5 Nano wins 11 of 11 shared benchmarks. Leads in math · knowledge · language.
Category leads
math·GPT-5 Nanoknowledge·GPT-5 Nanolanguage·GPT-5 Nanoreasoning·GPT-5 Nanocoding·GPT-5 Nano
Hype vs Reality
Attention vs performance
Claude 3.5 Sonnet
#129 by perf·no signal
GPT-5 Nano
#114 by perf·no signal
Vendor risk
Who is behind the model
Anthropic
$380.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
11 benchmarks · 2 models
Claude 3.5 SonnetGPT-5 Nano
FrontierMath-2025-02-28-Private
GPT-5 Nano leads by +7.2
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
Claude 3.5 Sonnet
1.0
GPT-5 Nano
8.3
FrontierMath-Tier-4-2025-07-01-Private
GPT-5 Nano leads by +2.0
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
Claude 3.5 Sonnet
0.1
GPT-5 Nano
2.1
GPQA diamond
GPT-5 Nano leads by +20.5
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude 3.5 Sonnet
38.7
GPT-5 Nano
59.3
HELM · GPQA
GPT-5 Nano leads by +11.4
Claude 3.5 Sonnet
56.5
GPT-5 Nano
67.9
HELM · IFEval
GPT-5 Nano leads by +7.6
Claude 3.5 Sonnet
85.6
GPT-5 Nano
93.2
HELM · MMLU-Pro
GPT-5 Nano leads by +0.1
Claude 3.5 Sonnet
77.7
GPT-5 Nano
77.8
HELM · Omni-MATH
GPT-5 Nano leads by +27.1
Claude 3.5 Sonnet
27.6
GPT-5 Nano
54.7
HELM · WildBench
GPT-5 Nano leads by +1.4
Claude 3.5 Sonnet
79.2
GPT-5 Nano
80.6
MATH level 5
GPT-5 Nano leads by +43.6
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Claude 3.5 Sonnet
51.7
GPT-5 Nano
95.2
OTIS Mock AIME 2024-2025
GPT-5 Nano leads by +74.7
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude 3.5 Sonnet
6.4
GPT-5 Nano
81.1
WeirdML
GPT-5 Nano leads by +7.1
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude 3.5 Sonnet
31.0
GPT-5 Nano
38.1
Full benchmark table
| Benchmark | Claude 3.5 Sonnet | GPT-5 Nano |
|---|---|---|
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. | 1.0 | 8.3 |
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. | 0.1 | 2.1 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 38.7 | 59.3 |
HELM · GPQA | 56.5 | 67.9 |
HELM · IFEval | 85.6 | 93.2 |
HELM · MMLU-Pro | 77.7 | 77.8 |
HELM · Omni-MATH | 27.6 | 54.7 |
HELM · WildBench | 79.2 | 80.6 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 51.7 | 95.2 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 6.4 | 81.1 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 31.0 | 38.1 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| $0.05 | $0.40 | 400K tokens (~200 books) | $1.38 |
People also compared