Tools
Compare AI models
Intelligence, speed and price side by side, so you can see what a model actually costs you for the quality it gives.
The smartest right now
| Model | Maker | Intelligence | Speed | Input / 1M | Output / 1M | Cost per point |
|---|---|---|---|---|---|---|
| Claude Opus 5.5 (Max, Default Fallback) | Anthropic | 57.6 | 97/s | $4 | $20 | $0.14 |
| Claude Sonnet 5.5 (Max, Default Fallback) | Anthropic | 56 | 142/s | $2 | $10 | $0.07 |
| Claude Opus 5.5 (Xhigh, Default Fallback) | Anthropic | 56 | 86/s | $4 | $20 | $0.14 |
| Muse Wonder Full (Max) | Meta | 55.8 | – | free | free | $0.00 |
| Claude Opus 5.5 (High, Default Fallback) | Anthropic | 53.6 | 79/s | $4 | $20 | $0.15 |
| Claude Fable 5.1 (Max, Default Fallback) | Anthropic | 53.4 | 73/s | $10 | $50 | $0.37 |
| Claude Fable 5.1 (Xhigh, Default Fallback) | Anthropic | 53.2 | 62/s | $10 | $50 | $0.38 |
| GPT-6 Astra (Max) | OpenAI | 52.7 | 69/s | $10 | $50 | $0.38 |
Best value: cheapest per point of intelligence
| Model | Maker | Intelligence | Speed | Input / 1M | Output / 1M | Cost per point |
|---|---|---|---|---|---|---|
| Muse Wonder Full (Max) | Meta | 55.8 | – | free | free | $0.00 |
| JT-4.1 Flash 236B A21B (Reasoning) | China Mobile | 33.9 | – | free | free | $0.00 |
| Motif 3 | Motif Technologies | 33.6 | – | free | free | $0.00 |
| Motif 3 (Beta) | Motif Technologies | 32.3 | – | free | free | $0.00 |
| Muse Spark | Meta | 31.3 | – | free | free | $0.00 |
| K2 Horizon 375B A23B | Institute of Foundation Models | 30.5 | 1065/s | free | free | $0.00 |
| MiMo-V2-Pro | Xiaomi | 28.6 | – | free | free | $0.00 |
| Qwen3.6 Max Preview | Alibaba | 28.4 | – | free | free | $0.00 |
Cheapest to run
| Model | Maker | Intelligence | Speed | Input / 1M | Output / 1M | Cost per point |
|---|---|---|---|---|---|---|
| Muse Wonder Full (Max) | Meta | 55.8 | – | free | free | $0.00 |
| JT-4.1 Flash 236B A21B (Reasoning) | China Mobile | 33.9 | – | free | free | $0.00 |
| Motif 3 | Motif Technologies | 33.6 | – | free | free | $0.00 |
| Motif 3 (Beta) | Motif Technologies | 32.3 | – | free | free | $0.00 |
| Muse Spark | Meta | 31.3 | – | free | free | $0.00 |
| K2 Horizon 375B A23B | Institute of Foundation Models | 30.5 | 1065/s | free | free | $0.00 |
| MiMo-V2-Pro | Xiaomi | 28.6 | – | free | free | $0.00 |
| Qwen3.6 Max Preview | Alibaba | 28.4 | – | free | free | $0.00 |
Fastest to answer
| Model | Maker | Intelligence | Speed | Input / 1M | Output / 1M | Cost per point |
|---|---|---|---|---|---|---|
| Celeris-1 | Celeris | 6.3 | 1770/s | $0.20 | $0.70 | $0.05 |
| K2 Horizon 375B A23B | Institute of Foundation Models | 30.5 | 1065/s | free | free | $0.00 |
| Mercury 2.5 | Inception | 12.3 | 878/s | $0.25 | $0.75 | $0.03 |
| Nemotron 3.5 Super VL | NVIDIA | 26.5 | 402/s | $0.23 | $0.57 | $0.01 |
| Gemini 3.5 Flash-Lite | 22.2 | 398/s | $0.30 | $2.50 | $0.04 | |
| Trinity Large Thinking | Arcee AI | 10.8 | 373/s | $0.25 | $0.80 | $0.04 |
| DeepSeek V4.1 Flash (Non-reasoning) | DeepSeek | 24.7 | 360/s | $0.30 | $1.20 | $0.02 |
| Ling-3.0-flash-Fin | InclusionAI | 22.6 | 302/s | $0.07 | $0.22 | $0.00 |
Data from Artificial Analysis, refreshed every few hours. Intelligence is their index built from several independent tests. Input is the text you send, output is what the model writes back, both priced per million tokens (a token is roughly three quarters of a word). Cost per point mixes the two prices three to one, the way real use does, and divides by the intelligence score: lower is better value.