15 models · avg score 49
Track reliability alongside pricing and model depth so provider selection stays grounded in operating reality.
Status monitoring via official APIs not yet available
Rolling availability across the last 30 days.
Composite reliability indicator from recent status behavior.
Includes degraded performance, partial outages, and major incidents.
| # | Model | Score |
|---|---|---|
| 100 | Llama 4 Maverick | 77 |
| 144 | Llama 4 Scout | 72 |
| 183 | Llama 3.3 70B Instruct | 66 |
| 224 | Llama 3.1 70B Instruct | 60 |
| 229 | Llama Guard 4 12B | 59 |
| 249 | Llama 3.2 11B Vision Instruct | 54 |
| 275 | Llama 3.3 70B Instruct (free) | 44 |
| 278 | Llama Guard 3 8B | 43 |
| 283 | Llama 3.1 8B Instruct | 42 |
| 285 | Llama 3 70B Instruct | 41 |
| 291 | Llama 3.1 405B (base) | 39 |
| 298 | Llama 3.2 3B Instruct | 36 |
| 299 | Llama 3.2 3B Instruct (free) | 35 |
| 303 | Llama 3.2 1B Instruct | 32 |
| 304 | Llama 3 8B Instruct | 30 |
| Rank | Model | Score |
|---|---|---|
| #100 | Llama 4 Maverick | 77 |
| #144 | Llama 4 Scout | 72 |
| #183 | Llama 3.3 70B Instruct | 66 |
| #224 | Llama 3.1 70B Instruct | 60 |
| #229 | Llama Guard 4 12B | 59 |
| #249 | Llama 3.2 11B Vision Instruct | 54 |
| #275 | Llama 3.3 70B Instruct (free) | 44 |
| #278 | Llama Guard 3 8B | 43 |
| #283 | Llama 3.1 8B Instruct | 42 |
| #285 | Llama 3 70B Instruct | 41 |
| #291 | Llama 3.1 405B (base) | 39 |
| #298 | Llama 3.2 3B Instruct | 36 |
| #299 | Llama 3.2 3B Instruct (free) | 35 |
| #303 | Llama 3.2 1B Instruct | 32 |
| #304 | Llama 3 8B Instruct | 30 |
Meta currently offers 15 AI models on our platform, with an average quality score of 49/100.
The highest-rated Meta model is Llama 4 Maverick, ranked #100 overall with a composite score of 77/100.
Yes, Meta offers 2 free models that you can use at no cost.
Meta models have an average score of 49/100. Their best model, Llama 4 Maverick, is ranked #100 globally. Visit our provider comparison page for detailed analysis.