Compare · ModelsLive · 2 picked · head to head
Claude 2.1 vs Claude 3.5 Haiku
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude 3.5 Haiku wins on 4/4 benchmarks
Claude 3.5 Haiku wins 4 of 4 shared benchmarks. Leads in knowledge · math · coding.
Category leads
knowledge·Claude 3.5 Haikumath·Claude 3.5 Haikucoding·Claude 3.5 Haiku
Hype vs Reality
Attention vs performance
Claude 2.1
#213 by perf·no signal
Claude 3.5 Haiku
#159 by perf·no signal
Vendor risk
Who is behind the model
Anthropic
$380.0B·Tier 1
Anthropic
$380.0B·Tier 1
Head to head
4 benchmarks · 2 models
Claude 2.1Claude 3.5 Haiku
GPQA diamond
Claude 3.5 Haiku leads by +6.9
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude 2.1
10.6
Claude 3.5 Haiku
17.5
MMLU
Claude 3.5 Haiku leads by +1.1
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Claude 2.1
64.7
Claude 3.5 Haiku
65.7
OTIS Mock AIME 2024-2025
Claude 3.5 Haiku leads by +2.4
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude 2.1
1.9
Claude 3.5 Haiku
4.2
WeirdML
Claude 3.5 Haiku leads by +23.7
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude 2.1
7.1
Claude 3.5 Haiku
30.7
Full benchmark table
| Benchmark | Claude 2.1 | Claude 3.5 Haiku |
|---|---|---|
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 10.6 | 17.5 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 64.7 | 65.7 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 1.9 | 4.2 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 7.1 | 30.7 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| $0.80 | $4.00 | 200K tokens (~100 books) | $16.00 |
People also compared