Llama series comparison
Meta's Llama family: open-source, self-hostable and commercially friendly, spanning many parameter sizes and long context.
All models
Llama series comparison
Meta's Llama family: open-source, self-hostable and commercially friendly, spanning many parameter sizes and long context.
6 versions (sorted by capability)
| Version | Context | Input | Output | Reasoning | Code | Speed | Open source |
|---|---|---|---|---|---|---|---|
| Llama 3.1 405B | 128K | 开源免费 | 开源免费 | ●●●●○ | ●●●●○ | ●●●○○ | Open |
| Llama 3.1 70B | 128K | 开源免费 | 开源免费 | ●●●○○ | ●●●○○ | ●●●●○ | Open |
| Llama 3.1 8B | 128K | 开源免费 | 开源免费 | ●●○○○ | ●●○○○ | ●●●●● | Open |
| Llama 3.3 70B | 128K | 开源免费 | 开源免费 | ●●●●○ | ●●●●○ | ●●●●○ | Open |
| Llama 4 Maverick | 1M | 开源免费 | 开源免费 | ●●●●○ | ●●●●○ | ●●●●○ | Open |
| Llama 4 Scout | 10M | 开源免费 | 开源免费 | ●●●●○ | ●●●○○ | ●●●●○ | Open |
Llama 3.1 405BMeta Llama 3.1 405B: A flagship open-source model with 405 billion parameters, 128K context, and performance approaching closed-source flagships.Llama 3.1 70BMeta Llama 3.1 70B: A 70-billion-parameter open-source general-purpose model with 128K context, offering high deployment cost-effectiveness.Llama 3.1 8BMeta Llama 3.1 8B: An 8-billion-parameter lightweight open-source model with 128K context, suitable for edge and low-cost deployment.Llama 3.3 70BMeta's latest open-source flagship, 70 billion parameters, self-hostable, business-friendly license, performance approaching closed-source models.Llama 4 MaverickMeta Llama 4 Maverick: 400B total params (17B activated, 128 experts) MoE, 1M context, open weights, commercially friendly.Llama 4 ScoutMeta's latest open-source model features a native 10M ultra-long context, supports self-hosting, and is commercially friendly.