llama2:70b-chat-q6_K

llama2

Llama 2 是一系列基础语言模型，参数量从 70 亿到 700 亿不等。

70亿 130亿 700亿

250万次下载更新于 11个月前

102个标签

11个月前更新

11个月前

b8bbc9a59804 · 57GB

[INST] <<SYS>>{{ .System }}<</SYS>> {{ .Prompt }} [/INST]

59B

参数

{ "stop": [ "[INST]", "[/INST]", "<<SYS>>", "<</SYS>>" ] }

91B

许可证

# Llama 2 Acceptable Use Policy Meta is committed to promoting safe and fair use of its tools and f

4.8kB

许可证

LLAMA 2 COMMUNITY LICENSE AGREEMENT Llama 2 Version Release Date: July 18, 2023 "Agreement" means

7.0kB

自述文件

Llama 2 由 Meta Platforms, Inc. 发布。该模型在 2 万亿个 token 上进行训练，默认支持 4096 的上下文长度。Llama 2 Chat 模型在超过 100 万个人工标注上进行了微调，专为聊天而设计。

命令行界面

打开终端并运行ollama run llama2

API

使用 curl 的示例

curl -X POST https://127.0.0.1:11434/api/generate -d '{
  "model": "llama2",
  "prompt":"Why is the sky blue?"
 }'

API 文档

内存需求

70亿参数模型通常至少需要 8GB 的 RAM
130亿参数模型通常至少需要 16GB 的 RAM
700亿参数模型通常至少需要 64GB 的 RAM

如果在更高的量化级别上遇到问题，请尝试使用 q4 模型或关闭任何其他占用大量内存的程序。

模型变体

**聊天** 模型针对聊天/对话用例进行了微调。这些是 Ollama 中的默认模型，以及在标签选项卡中标有 -chat 的模型。

示例：ollama run llama2

**预训练** 模型没有聊天微调。这些在标签选项卡中标有 -text。

示例：ollama run llama2:text

默认情况下，Ollama 使用 4 位量化。要尝试其他量化级别，请尝试其他标签。q 后的数字表示用于量化的位数（例如，q4 表示 4 位量化）。数字越大，模型越准确，但运行速度越慢，需要的内存也越多。

参考文献

Llama 2：开放式基础和微调聊天模型

Meta 的 Hugging Face 代码库

Llama 2 is released by Meta Platforms, Inc. This model is trained on 2 trillion tokens, and by default supports a context length of 4096. Llama 2 Chat models are fine-tuned on over 1 million human annotations, and are made for chat.

### CLI

Open the terminal and run `ollama run llama2`

### API

Example using curl:

```bash
curl -X POST https://127.0.0.1:11434/api/generate -d '{
  "model": "llama2",
  "prompt":"Why is the sky blue?"
 }'
```

[API documentation](https://github.com/jmorganca/ollama/blob/main/docs/api.md)

## Memory requirements

- 7b models generally require at least 8GB of RAM
- 13b models generally require at least 16GB of RAM
- 70b models generally require at least 64GB of RAM

If you run into issues with higher quantization levels, try using the q4 model or shut down any other programs that are using a lot of memory.

## Model variants

**Chat** is fine-tuned for chat/dialogue use cases. These are the default in Ollama, and for models tagged with -chat in the tags tab.

*Example: `ollama run llama2`*

**Pre-trained** is without the chat fine-tuning. This is tagged as -text in the tags tab.

*Example: `ollama run llama2:text`*

By default, Ollama uses 4-bit quantization. To try other quantization levels, please try the other tags. The number after the q represents the number of bits used for quantization (i.e. q4 means 4-bit quantization). The higher the number, the more accurate the model is, but the slower it runs, and the more memory it requires.

## References
[Llama 2: Open Foundation and Fine-Tuned Chat Models](https://arxiv.org/abs/2307.09288)

[Meta's Hugging Face repo](https://hugging-face.cn/meta-llama)

粘贴、拖放或点击上传图像（.png、.jpeg、.jpg、.svg、.gif）