Best Meta Llama model overall
- models ranked
- 7
- lead over #2
- +1.6
Overall index · read October 21
As of October 21, 2026, Meta Llama 4 Maverick leads 7 models that qualify, with 9.3 on the overall index, ahead of Meta Llama 3.3 70B Instruct (7.7) and Meta Llama 3.1 8B Instruct (6.9).
| Rank | Model | Overall index | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Llama 4 Maverick Meta Llama | 9.3 | 18.1 | $0.6525 | |
| 2 | Llama 3.3 70B Instruct Meta Llama | 7.7 | Not benchmarked | $0.32 | |
| 3 | Llama 3.1 8B Instruct Meta Llama | 6.9 | Not benchmarked | $0.08 | |
| 4 | Llama 3.1 70B Instruct Meta Llama | 6.6 | Not benchmarked | $0.40 | |
| 5 | Llama 4 Scout Meta Llama | 6.5 | 9.1 | $0.30 | |
| 6 | Llama 3.2 3B Instruct Meta Llama | 5.7 | Not benchmarked | $0.335 | |
| 7 | Llama 3.2 1B Instruct Meta Llama | 4.8 | Not benchmarked | $0.201 |
How this list is built
Generated from the dataset on every refresh. Nobody picks the order by hand.
Every Meta Llama model on the overall index, ordered by the Artificial Analysis Intelligence Index, highest first. The index is a composite of evaluations Artificial Analysis runs itself, published on a 0 to 100 scale.
- OpenRouter, fetched
- Artificial Analysis (via OpenRouter), fetched