Meta cloud models
18 models tracked via Artificial Analysis. Compare cloud performance, then find local GGUF versions in the GraySoft model catalog.
| Model | Intelligence | Speed (tok/s) |
|---|---|---|
| Muse Spark 1.1 (xhigh) | 50.6 | 137.405 |
| Muse Spark | 43.1 | 0 |
| Llama 4 Maverick | 14.3 | 121.692 |
| Llama 4 Scout | 10 | 97.698 |
| Llama 3.3 Instruct 70B | 9.4 | 86.352 |
| Llama 3.1 Instruct 405B | 8.5 | 0 |
| Llama 3.1 Instruct 8B | 7.6 | 614.701 |
| Llama 3.1 Instruct 70B | 6.8 | 0 |
| Llama 3.2 Instruct 90B (Vision) | 6.2 | 0 |
| Llama 2 Chat 7B | 4.3 | 0 |
| Llama 3.2 Instruct 3B | 4.2 | 0 |
| Llama 3 Instruct 70B | 3.5 | 0 |
| Llama 3.2 Instruct 11B (Vision) | 3.3 | 8.336 |
| Llama 2 Chat 70B | 3 | 0 |
| Llama 2 Chat 13B | 3 | 0 |
| Llama 65B | 2.1 | 0 |
| Llama 3 Instruct 8B | 1.2 | 0 |
| Llama 3.2 Instruct 1B | 1.1 | 0 |
Run models locally with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.