Compare · ModelsLive · 3 picked · head to head
Claude 3 Opus vs GPT-4 Turbo (batch) vs Llama 3.1 405B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-4 Turbo (batch) wins on 6/14 benchmarks
GPT-4 Turbo (batch) wins 6 of 14 shared benchmarks. Leads in general · reasoning.
Category leads
general·GPT-4 Turbo (batch)knowledge·Llama 3.1 405Bmath·Llama 3.1 405Breasoning·GPT-4 Turbo (batch)coding·Llama 3.1 405B
Hype vs Reality
Attention vs performance
Claude 3 Opus
#276 by perf·no signal
GPT-4 Turbo (batch)
#190 by perf·no signal
Llama 3.1 405B
#222 by perf·no signal
Best value
GPT-4 Turbo (batch)
Claude 3 Opus
n/a
no price
GPT-4 Turbo (batch)
4.2 pts/$
$10.00/M
Llama 3.1 405B
n/a
no price
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
OpenAI
$840.0B·Tier 1
Meta AI
$1.62T·Tier 1
Head to head
14 benchmarks · 3 models
Claude 3 OpusGPT-4 Turbo (batch)Llama 3.1 405B
Dtbench
GPT-4 Turbo (batch) leads by +1.8
Claude 3 Opus
36.0
GPT-4 Turbo (batch)
37.8
Llama 3.1 405B
35.6
GPQA diamond
Llama 3.1 405B leads by +5.2
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
30.8
GPT-4 Turbo (batch)
7.8
Llama 3.1 405B
36.0
MATH level 5
Llama 3.1 405B leads by +12.3
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Claude 3 Opus
37.5
GPT-4 Turbo (batch)
23.0
Llama 3.1 405B
49.8
MMLU
Claude 3 Opus leads by +0.1
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Claude 3 Opus
79.5
GPT-4 Turbo (batch)
76.5
Llama 3.1 405B
79.3
OTIS Mock AIME 2024-2025
Llama 3.1 405B leads by +5.0
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude 3 Opus
4.6
GPT-4 Turbo (batch)
1.0
Llama 3.1 405B
9.6
SimpleBench
GPT-4 Turbo (batch) leads by +1.9
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude 3 Opus
8.2
GPT-4 Turbo (batch)
10.1
Llama 3.1 405B
7.6
WeirdML
Llama 3.1 405B leads by +2.2
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude 3 Opus
19.2
GPT-4 Turbo (batch)
12.4
Llama 3.1 405B
21.4
Winogrande
Llama 3.1 405B leads by +1.4
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
Claude 3 Opus
77.0
GPT-4 Turbo (batch)
75.0
Llama 3.1 405B
78.4
BBH
Llama 3.1 405B leads by +10.4
BIG-Bench Hard · a curated subset of 23 challenging tasks from BIG-Bench where language models previously failed to outperform average humans.
GPT-4 Turbo (batch)
66.8
Llama 3.1 405B
77.2
Cybench
Claude 3 Opus leads by +2.5
Cybench · evaluates AI on real Capture-The-Flag cybersecurity challenges, testing vulnerability analysis, exploitation, and security reasoning.
Claude 3 Opus
10.0
Llama 3.1 405B
7.5
HellaSwag
GPT-4 Turbo (batch) leads by +8.1
HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios.
GPT-4 Turbo (batch)
93.7
Llama 3.1 405B
85.6
Lmca
GPT-4 Turbo (batch) leads by +0.1
Claude 3 Opus
20.0
GPT-4 Turbo (batch)
20.1
Metr Time Horizons
GPT-4 Turbo (batch) leads by +6.6
Claude 3 Opus
29.5
GPT-4 Turbo (batch)
36.1
TriviaQA
GPT-4 Turbo (batch) leads by +2.1
TriviaQA · reading comprehension benchmark with trivia questions, requiring models to find and reason over evidence from provided documents.
GPT-4 Turbo (batch)
84.8
Llama 3.1 405B
82.7
Full benchmark table
| Benchmark | Claude 3 Opus | GPT-4 Turbo (batch) | Llama 3.1 405B |
|---|---|---|---|
Dtbench | 36.0 | 37.8 | 35.6 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 30.8 | 7.8 | 36.0 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 37.5 | 23.0 | 49.8 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 79.5 | 76.5 | 79.3 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 4.6 | 1.0 | 9.6 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 8.2 | 10.1 | 7.6 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 19.2 | 12.4 | 21.4 |
Winogrande WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs. | 77.0 | 75.0 | 78.4 |
BBH BIG-Bench Hard · a curated subset of 23 challenging tasks from BIG-Bench where language models previously failed to outperform average humans. | n/a | 66.8 | 77.2 |
Cybench Cybench · evaluates AI on real Capture-The-Flag cybersecurity challenges, testing vulnerability analysis, exploitation, and security reasoning. | 10.0 | n/a | 7.5 |
HellaSwag HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios. | n/a | 93.7 | 85.6 |
Lmca | 20.0 | 20.1 | n/a |
Metr Time Horizons | 29.5 | 36.1 | n/a |
TriviaQA TriviaQA · reading comprehension benchmark with trivia questions, requiring models to find and reason over evidence from provided documents. | n/a | 84.8 | 82.7 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| n/a | n/a | n/a | n/a | |
| $5.00 | $15.00 | 128K tokens (~64 books) | $75.00 | |
| n/a | n/a | n/a | n/a |