Compare · ModelsLive · 2 picked · head to head
Claude 3 Haiku vs Llama 2-13B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude 3 Haiku wins on 7/7 benchmarks
Claude 3 Haiku wins 7 of 7 shared benchmarks. Leads in general · knowledge · math.
Category leads
general·Claude 3 Haikuknowledge·Claude 3 Haikumath·Claude 3 Haiku
Hype vs Reality
Attention vs performance
Claude 3 Haiku
#271 by perf·no signal
Llama 2-13B
#224 by perf·#18 by attention
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
Meta AI
$1.87T·Tier 1
Head to head
7 benchmarks · 2 models
Claude 3 HaikuLlama 2-13B
Dtbench
Claude 3 Haiku leads by +13.1
Claude 3 Haiku
16.9
Llama 2-13B
3.7
GPQA diamond
Claude 3 Haiku leads by +13.3
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude 3 Haiku
15.1
Llama 2-13B
1.8
MATH level 5
Claude 3 Haiku leads by +11.6
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Claude 3 Haiku
14.9
Llama 2-13B
3.3
MMLU
Claude 3 Haiku leads by +24.3
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Claude 3 Haiku
65.1
Llama 2-13B
40.8
OTIS Mock AIME 2024-2025
Claude 3 Haiku leads by +1.7
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude 3 Haiku
1.7
Llama 2-13B
0.0
ScienceQA
Claude 3 Haiku leads by +21.6
ScienceQA · multimodal science questions spanning natural science, social science, and language science with diverse question formats and image context.
Claude 3 Haiku
62.7
Llama 2-13B
41.0
Winogrande
Claude 3 Haiku leads by +2.8
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
Claude 3 Haiku
48.4
Llama 2-13B
45.6
Full benchmark table
| Benchmark | Claude 3 Haiku | Llama 2-13B |
|---|---|---|
Dtbench | 16.9 | 3.7 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 15.1 | 1.8 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 14.9 | 3.3 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 65.1 | 40.8 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 1.7 | 0.0 |
ScienceQA ScienceQA · multimodal science questions spanning natural science, social science, and language science with diverse question formats and image context. | 62.7 | 41.0 |
Winogrande WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs. | 48.4 | 45.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.25 | $1.25 | 200K tokens (~100 books) | $5.00 | |
| — | — | — | — |