Model Comparator
Compare up to four models using the same criteria.
2 models compared
- GPT-5.4 mini (Xhigh)
OpenAI
- Claude Opus 4.5 (Non-reasoning)
Anthropic
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | GPT-5.4 mini (Xhigh) OpenAI | Claude Opus 4.5 (Non-reasoning) Anthropic | Add a model · 2/4 |
|---|---|---|---|
| Intelligence | Best: 24.1 | 23.7 | |
| Coding | 56.1 | — | |
| Math | — | 62.7 | |
| Agentic | 17.9 | — | |
| Speed | Best: 233 t/s | 46.1 t/s | |
| TTFT | 114 s | Best: 1.19 s | |
| Input · $ / 1M tokens | Best: $0.75 | $5×6.7 relative to the lowest price among the compared models | |
| Output · $ / 1M tokens | Best: $4.50 | $25×5.6 relative to the lowest price among the compared models | |
| Context window | Best: 400K | 200K | |
| Reasoning | Yes | No | |
| Open weights | No | No | |
| Cost / task | $0.447 | — |
Intelligence
- GPT-5.4 mini (Xhigh)
- Best: 24.1
- Claude Opus 4.5 (Non-reasoning)
- 23.7
Coding
- GPT-5.4 mini (Xhigh)
- 56.1
- Claude Opus 4.5 (Non-reasoning)
- —
Math
- GPT-5.4 mini (Xhigh)
- —
- Claude Opus 4.5 (Non-reasoning)
- 62.7
Agentic
- GPT-5.4 mini (Xhigh)
- 17.9
- Claude Opus 4.5 (Non-reasoning)
- —
Speed
- GPT-5.4 mini (Xhigh)
- Best: 233 t/s
- Claude Opus 4.5 (Non-reasoning)
- 46.1 t/s
TTFT
- GPT-5.4 mini (Xhigh)
- 114 s
- Claude Opus 4.5 (Non-reasoning)
- Best: 1.19 s
Input · $ / 1M tokens
- GPT-5.4 mini (Xhigh)
- Best: $0.75
- Claude Opus 4.5 (Non-reasoning)
- $5×6.7 relative to the lowest price among the compared models
Output · $ / 1M tokens
- GPT-5.4 mini (Xhigh)
- Best: $4.50
- Claude Opus 4.5 (Non-reasoning)
- $25×5.6 relative to the lowest price among the compared models
Context window
- GPT-5.4 mini (Xhigh)
- Best: 400K
- Claude Opus 4.5 (Non-reasoning)
- 200K
Reasoning
- GPT-5.4 mini (Xhigh)
- Yes
- Claude Opus 4.5 (Non-reasoning)
- No
Open weights
- GPT-5.4 mini (Xhigh)
- No
- Claude Opus 4.5 (Non-reasoning)
- No
Cost / task
- GPT-5.4 mini (Xhigh)
- $0.447
- Claude Opus 4.5 (Non-reasoning)
- —
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.
- GPT-5.4 mini (Xhigh)24.1 · $1.69
- Claude Opus 4.5 (Non-reasoning)23.7 · $10