nous-hermes:13b-llama2-q4_1

curl -X POST https://127.0.0.1:11434/api/generate -d '{
  "model": "nous-hermes",
  "prompt":"Explain the process of how a refrigerator works to keep the contents inside cold."
 }'

CLI

安裝 Ollama
開啟終端機並執行 ollama run nous-hermes

注意：如果模型尚未下載，ollama run 命令會執行 ollama pull。若要在不執行的情況下下載模型，請使用 ollama pull nous-hermes

記憶體需求

7b 模型通常至少需要 8GB 的 RAM
13b 模型通常至少需要 16GB 的 RAM

如果您在使用較高的量化層級時遇到問題，請嘗試使用 q4 模型，或關閉任何其他佔用大量記憶體的程式。

模型變體

Ollama 提供了許多 Nous Hermes 模型的變體，這些變體基於官方模型進行量化，以便在本地良好運行。

Nous Hermes Llama 2 是基於原始 Llama 模型的原始 Nous Hermes 模型。

範例：ollama run nous-hermes

Nous Hermes Llama 1 是基於原始 Llama 模型的原始 Nous Hermes 模型。

範例：ollama run nous-hermes:13b-q4_0

預設情況下，Ollama 使用 4 位元量化。若要嘗試其他量化層級，請嘗試其他標籤。q 後面的數字代表用於量化的位元數 (即 q4 表示 4 位元量化)。數字越高，模型越準確，但執行速度越慢，且需要的記憶體越多。

別名
latest, 7b, 7b-llama2, 7b-llama2-q4_0
13b, 13b-llama2, 13b-llama2-q4_0

模型來源

Ollama 上的 Nous Hermes Llama 2 來源

7b 參數原始來源：Nous Research

13b 參數原始來源：Nous Research

Ollama 上的 Nous Hermes Llama 1 來源

13b 參數原始來源：Nous Research

Nous Hermes was released by Nous Research. There are two main variants here, a 13B parameter model based on Llama, and a 7B and 13B parameter model based on Llama 2. They are all general-use models trained with the same datasets.

## Get started with Nous Hermes

The model used in the example below is the Nous Hermes Llama 2 model, with 7b parameters, which is a general chat model.

### API

1. Start Ollama server (Run `ollama serve`)
2. Run the model

```bash
curl -X POST https://127.0.0.1:11434/api/generate -d '{
  "model": "nous-hermes",
  "prompt":"Explain the process of how a refrigerator works to keep the contents inside cold."
 }'
  ```

### CLI

1. Install Ollama
2. Open the terminal and run `ollama run nous-hermes`

Note: The `ollama run` command performs an `ollama pull` if the model is not already downloaded. To download the model without running it, use `ollama pull nous-hermes`

## Memory requirements

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

Ollama offers many variants of the Nous Hermes model that are quantized based on the official models to run well locally.

**Nous Hermes Llama 2** is the original Nous Hermes model based on the original Llama model.

*Example: `ollama run nous-hermes`*

**Nous Hermes Llama 1** is the original Nous Hermes model based on the original Llama model.

*Example: `ollama run nous-hermes:13b-q4_0`*

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.

| **Aliases** |
| --- |
| latest, 7b, 7b-llama2, 7b-llama2-q4_0 |
| 13b, 13b-llama2, 13b-llama2-q4_0 |

## Model source

**Nous Hermes Llama 2 source on Ollama**

7b parameters original source:
 [Nous Research](https://huggingface.co/NousResearch/Nous-Hermes-llama-2-7b)

13b parameters original source:
 [Nous Research](https://huggingface.co/NousResearch/Nous-Hermes-Llama2-13b)

**Nous Hermes Llama 1 source on Ollama**

13b parameters original source:
 [Nous Research](https://huggingface.co/NousResearch/Nous-Hermes-13b)

貼上、拖曳或點擊以上傳圖片 (.png, .jpeg, .jpg, .svg, .gif)