Tools
Choose what you want the model to do. We automatically use the right live benchmark.
Balanced capability for assistants and everyday tasks. These results come from the current database and the benchmark selected for your task.
| Rank | Model | Best for | Result | Approx. API price |
|---|---|---|---|---|
| #1 | Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic | Everyday tasks | 59.9 | $20 / 1M |
| #2 | GPT-5.6 Sol (max)OpenAI | Everyday tasks | 58.9 | $11.25 / 1M |
| #3 | GPT-5.6 Sol (xhigh)OpenAI | Everyday tasks | 57.7 | $11.25 / 1M |
| #4 |
You do not need to understand benchmark jargon. Start with what you want to achieve.
Start with coding, math, speed, price, or general capability.
See a short ranking calculated from the current benchmark database.
Open the exact result page and methodology before you decide.
| GPT-5.6 Sol (high)OpenAI |
| Everyday tasks |
| 55.9 |
| $11.25 / 1M |
| #5 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic | Everyday tasks | 55.7 | $10 / 1M |
| #6 | GPT-5.6 Terra (max)OpenAI | Everyday tasks | 55.0 | $5.625 / 1M |
| #7 | GPT-5.5 (xhigh)OpenAI | Everyday tasks | 54.8 | $11.25 / 1M |
| #8 | Grok 4.5 (high)SpaceXAI | Everyday tasks | 53.8 | $3 / 1M |
| #9 | GPT-5.6 Sol (medium)OpenAI | Everyday tasks | 53.6 | $11.25 / 1M |
| #10 | Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic | Everyday tasks | 53.5 | $10 / 1M |