Model Comparator
Compare up to four models using the same criteria.
2 models compared
- Claude 4.5 Haiku (Reasoning)
Anthropic
- GPT-5 mini (High)
OpenAI
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | Claude 4.5 Haiku (Reasoning) Anthropic | GPT-5 mini (High) OpenAI | Add a model · 2/4 |
|---|---|---|---|
| Intelligence | Best: 16.9 | 16.8 | |
| Coding | Best: 43.9 | 15.6 | |
| Math | 83.7 | Best: 90.7 | |
| Agentic | Best: 8 | 6.8 | |
| Speed | 85.9 t/s | Best: 147 t/s | |
| TTFT | Best: 20.8 s | 49.5 s | |
| Input · $ / 1M tokens | $1×4 relative to the lowest price among the compared models | Best: $0.25 | |
| Output · $ / 1M tokens | $5×2.5 relative to the lowest price among the compared models | Best: $2 | |
| Context window | 200K | Best: 400K | |
| Reasoning | Yes | Yes | |
| Open weights | No | No | |
| Cost / task | $0.277×5.2 relative to the lowest price among the compared models | Best: $0.0535 |
Intelligence
- Claude 4.5 Haiku (Reasoning)
- Best: 16.9
- GPT-5 mini (High)
- 16.8
Coding
- Claude 4.5 Haiku (Reasoning)
- Best: 43.9
- GPT-5 mini (High)
- 15.6
Math
- Claude 4.5 Haiku (Reasoning)
- 83.7
- GPT-5 mini (High)
- Best: 90.7
Agentic
- Claude 4.5 Haiku (Reasoning)
- Best: 8
- GPT-5 mini (High)
- 6.8
Speed
- Claude 4.5 Haiku (Reasoning)
- 85.9 t/s
- GPT-5 mini (High)
- Best: 147 t/s
TTFT
- Claude 4.5 Haiku (Reasoning)
- Best: 20.8 s
- GPT-5 mini (High)
- 49.5 s
Input · $ / 1M tokens
- Claude 4.5 Haiku (Reasoning)
- $1×4 relative to the lowest price among the compared models
- GPT-5 mini (High)
- Best: $0.25
Output · $ / 1M tokens
- Claude 4.5 Haiku (Reasoning)
- $5×2.5 relative to the lowest price among the compared models
- GPT-5 mini (High)
- Best: $2
Context window
- Claude 4.5 Haiku (Reasoning)
- 200K
- GPT-5 mini (High)
- Best: 400K
Reasoning
- Claude 4.5 Haiku (Reasoning)
- Yes
- GPT-5 mini (High)
- Yes
Open weights
- Claude 4.5 Haiku (Reasoning)
- No
- GPT-5 mini (High)
- No
Cost / task
- Claude 4.5 Haiku (Reasoning)
- $0.277×5.2 relative to the lowest price among the compared models
- GPT-5 mini (High)
- Best: $0.0535
Each reasoning level remains a separate configuration.
Score and cost
Each line joins the reasoning levels of a compared model; the large point marks the configuration in the table. The top-right corner holds the best score for the lowest cost. Select another point to compare that level.
- Claude 4.5 Haiku (Reasoning)16.9 · $0.277
- GPT-5 mini (High)16.8 · $0.0535