Llama 3.1 Family — Meta 的本地 AI 模型
作者： Jakub Rusinowski · 最后更新：
消费级硬件开源权重的领军者。Llama 3.1拥有高达128k的上下文窗口，在众多基准测试中媲美GPT-4o。
变体
Llama 3.1 Family 最小的变体在 Q4_K_M 下约需 6 GB 显存——量化权重加框架开销，不含 KV 缓存。
| 模型 | Q4 下显存 | 显存 | 上下文 | 运行 |
|---|---|---|---|---|
| Llama 3.1 8B Instruct → 8B | ~5.6 GB | 128,000 | ollama run llama3.1 |
显存为 Q4_K_M 下的量化权重加开销，与 GPU 与显存检测器使用同一引擎计算。
如何在本地运行 Llama 3.1 Family
安装 Ollama，然后拉取标签。
ollama run llama3.1在上方选择一个尺寸，查看它自己的显存、速度估算和安装命令。
Download Llama 3.1 Family — GGUF Quantizations
Pick a quantization and open it in LM Studio, Ollama, or Jan, or download the raw .gguf file directly. Quant list and sizes resolved from Hugging Face.
Llama 3.1 8B Instruct — GGUF quants · bartowski/Meta-Llama-3.1-8B-Instruct-GGUF
| Quant | Size | Download (.gguf) |
|---|---|---|
| Q3_K_M | 3.41 GB (est.) | Meta-Llama-3.1-8B-Instruct-Q3_K_M.gguf |
| Q4_K_M | 4.83 GB (est.) | Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf |
| Q5_K_M | 5.67 GB (est.) | Meta-Llama-3.1-8B-Instruct-Q5_K_M.gguf |
| Q6_K | 6.56 GB (est.) | Meta-Llama-3.1-8B-Instruct-Q6_K.gguf |
| Q8_0 | 8.50 GB (est.) | Meta-Llama-3.1-8B-Instruct-Q8_0.gguf |
Download in LM Studio: lms get bartowski/Meta-Llama-3.1-8B-Instruct-GGUF
Want this model on your phone? You can run it on your desktop with LM Studio and chat from your iPhone or iPad over an encrypted link — see Run LM Studio Models on Your Phone (LM Link).
许可证
Weights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms.
适用于: Llama 3.1 8B Instruct推荐 GPU
目录中能在本地运行 Llama 3.1 Family（至少 6 GB 显存）的最便宜 GPU 是 Intel Arc B570 (10 GB).
我的 GPU 能运行 Llama 3.1 Family 吗？
- Llama 3.1 Family 在 AMD Radeon RX 9060 XT 8GB 上
- Llama 3.1 Family 在 Intel Arc B570 上
- Llama 3.1 Family 在 Intel Arc B580 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 3060 (12GB) 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 3070 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 3070 Ti 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 3080 (10GB) 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 4060 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 4070 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 4070 Super 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 4070 Ti 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 5060 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 5060 Ti 8GB 上
- Llama 3.1 Family 在 NVIDIA GeForce RTX 5070 上
Llama 3.1 Family — 常见问题
How much VRAM does Llama 3.1 Family need?
Llama 3.1 Family needs about 6 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Llama 3.1 8B Instruct (6 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run Llama 3.1 Family on an RTX 4090 (24 GB)?
Yes — Llama 3.1 Family runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for Llama 3.1 Family?
Q4_K_M is the best balance of quality and VRAM for Llama 3.1 Family in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run Llama 3.1 Family with Ollama?
Install Ollama, then run: ollama run llama3.1. This downloads Llama 3.1 Family and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.