Claude 3.5 Haiku vs DeepSeek V3.2 Speciale
Anthropic vs DeepSeek — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 54.5)
- •DeepSeek V3.2 Speciale is cheaper ($0.29 vs $0.80 per 1M input)
- •Claude 3.5 Haiku has a larger context window (200K)
| Claude 3.5 Haiku | DeepSeek V3.2 Speciale | |
|---|---|---|
| Intelligence index | 54.5 | 89.9 |
| Developer | Anthropic | DeepSeek |
| Type | LLM | LLM |
| Access | API only | Open weights |
| Context window | 200,000 tokens | 163,840 tokens |
| Input price | $0.80 / 1M | $0.29 / 1M |
| Output price | $4.00 / 1M | $0.43 / 1M |
| Speed | 104 tok/s | — |
| Released | November 4, 2024 | December 1, 2025 |
| Parameters | — | — |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
Shared benchmarks
Claude 3.5 Haiku
DeepSeek V3.2 Speciale
GPQA Diamond
41.6
87.1
Humanity’s Last Exam
3.5
26.1
LiveCodeBench
31.4
89.6
MMLU-Pro
65
86.3