Compare · ModelsLive · 2 picked · head to head
GPT-4.1 Nano vs Llama 4 Maverick
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Llama 4 Maverick wins on 8/11 benchmarks
Llama 4 Maverick wins 8 of 11 shared benchmarks. Leads in coding · reasoning · general.
Category leads
coding·Llama 4 Maverickreasoning·Llama 4 Maverickgeneral·Llama 4 Maverickknowledge·Llama 4 Maverickmath·GPT-4.1 Nano
Hype vs Reality
Attention vs performance
GPT-4.1 Nano
#246 by perf·no signal
Llama 4 Maverick
#262 by perf·#18 by attention
Best value
GPT-4.1 Nano
1.9x better value than Llama 4 Maverick
GPT-4.1 Nano
124.0 pts/$
$0.25/M
Llama 4 Maverick
66.4 pts/$
$0.42/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Meta AI
$1.87T·Tier 1
Head to head
11 benchmarks · 2 models
GPT-4.1 NanoLlama 4 Maverick
Aider polyglot
Llama 4 Maverick leads by +6.7
Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.
GPT-4.1 Nano
8.9
Llama 4 Maverick
15.6
ARC-AGI
Llama 4 Maverick leads by +4.4
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
GPT-4.1 Nano
0.0
Llama 4 Maverick
4.4
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.
GPT-4.1 Nano
0.0
Llama 4 Maverick
0.0
Dtbench
Llama 4 Maverick leads by +15.6
GPT-4.1 Nano
20.9
Llama 4 Maverick
36.5
Fiction.LiveBench
Llama 4 Maverick leads by +21.2
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
GPT-4.1 Nano
25.0
Llama 4 Maverick
46.2
FrontierMath-2025-02-28-Private
GPT-4.1 Nano leads by +0.6
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GPT-4.1 Nano
1.8
Llama 4 Maverick
1.2
GPQA diamond
Llama 4 Maverick leads by +24.1
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-4.1 Nano
31.9
Llama 4 Maverick
56.0
Lmca
Llama 4 Maverick leads by +12.3
GPT-4.1 Nano
6.4
Llama 4 Maverick
18.7
MATH level 5
Llama 4 Maverick leads by +3.0
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
GPT-4.1 Nano
70.0
Llama 4 Maverick
73.0
OTIS Mock AIME 2024-2025
GPT-4.1 Nano leads by +8.3
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-4.1 Nano
28.8
Llama 4 Maverick
20.5
WeirdML
Llama 4 Maverick leads by +5.5
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
GPT-4.1 Nano
19.0
Llama 4 Maverick
24.5
Full benchmark table
| Benchmark | GPT-4.1 Nano | Llama 4 Maverick |
|---|---|---|
Aider polyglot Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework. | 8.9 | 15.6 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 0.0 | 4.4 |
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. | 0.0 | 0.0 |
Dtbench | 20.9 | 36.5 |
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 25.0 | 46.2 |
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. | 1.8 | 1.2 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 31.9 | 56.0 |
Lmca | 6.4 | 18.7 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 70.0 | 73.0 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 28.8 | 20.5 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 19.0 | 24.5 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.10 | $0.40 | 1.0M tokens (~524 books) | $1.75 | |
| $0.19 | $0.65 | 1.0M tokens (~524 books) | $3.04 |