All Comparisons
Claude Haiku 4.5 vs o3
Claude Haiku 4.5
Anthropic 🇺🇸
58.7
#157—
medium$1.00/M in · $5.00/M out
200K context
o3
OpenAI 🇺🇸
47.4
#202—
highMost Affordable
Claude Haiku 4.5
$1.00/M input
Largest Context
Claude Haiku 4.5
200K tokens
Best Benchmark Score
Claude Haiku 4.5
58.7/100
Model Specifications
| Spec | Claude Haiku 4.5 | o3 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Release Date | Oct 2025 | Apr 2025 |
| Knowledge Cutoff | 2025-02-01 | 2024-05-31 |
| Parameters | Undisclosed | Undisclosed |
| Context Window | 200K | — |
| Max Output | — | — |
| Open Source | No | No |
| License | proprietary | proprietary |
| Tokenizer | — | — |
| Modality | — | — |
| Reasoning Model | No | No |
| Moderated | No | No |
Pricing Comparison
| Price (per 1M tokens) | Claude Haiku 4.5 | o3 |
|---|---|---|
| Input | $1.00 | — |
| Output | $5.00 | — |
API Provider Pricing
Capabilities
| Feature | Claude Haiku 4.5 | o3 |
|---|---|---|
| Text Input | ||
| Image Input (Vision) | ||
| Audio Input | ||
| Video Input | ||
| File Input | ||
| Image Output | ||
| Audio Output | ||
| Tool Use | ||
| Structured Output (JSON) | ||
| Streaming | ||
| Reasoning Tokens | ||
| Web Search | ||
| Temperature Control | ||
| Top-P Sampling | ||
| Stop Sequences | ||
| Seed (Reproducibility) | ||
| Log Probabilities |
Category Comparison
Benchmark-by-Benchmark
general
| Benchmark | Claude Haiku 4.5 | o3 |
|---|---|---|
| MMMLU | 83 | — |
coding
| Benchmark | Claude Haiku 4.5 | o3 |
|---|---|---|
| SWE-bench Verified | 73.3 | 69.1 |
| TerminalBench | 41 | — |
math
| Benchmark | Claude Haiku 4.5 | o3 |
|---|---|---|
| AIME 2025 | 80.7 | 86.4 |
| FrontierMath | — | 15.8 |
reasoning
| Benchmark | Claude Haiku 4.5 | o3 |
|---|---|---|
| GPQA Diamond | 73 | 83.3 |
| Humanity's Last Exam | — | 14.7 |
| ARC-AGI v2 | — | 6.5 |
agent
| Benchmark | Claude Haiku 4.5 | o3 |
|---|---|---|
| OSWorld | 50.7 | — |
| BrowseComp | — | 49.7 |
multimodal
| Benchmark | Claude Haiku 4.5 | o3 |
|---|---|---|
| MMMU | — | 82.9 |
| MMMU-Pro | — | 76.4 |
| CharXiv Reasoning | — | 78.6 |
Category Winners
coding
70.2
math
53.4
reasoning
59.1
general
48.8
multimodal
72.1
agent
57.3