# Using DeepSeek locally and VScode extension

**URL:** <https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926>\
**Category:** Tutorials & Methods\
**Tags:** tips-tricks, ai\
**Created:** [January 28, 2025, 8:30am UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926 "2025-01-28T08:30:02Z")\
**Posts on this page:** 10\
**Page:** 1

<div class="post-metadata">

**Author:** ![Pratik\_Patnaik](https://onehack.st/user_avatar/onehack.st/pratik_patnaik/32/91937_2.png) [@Pratik\_Patnaik](https://onehack.st/u/Pratik_Patnaik)\
**Post date:** [January 28, 2025, 8:30am UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/1 "2025-01-28T08:30:03Z")

</div>

### **How to Use DeepSeek with Ollama and Continue in VSCode**

**A Step-by-Step Tutorial**

#### **Prerequisites**

1. **VSCode** installed on your machine.
2. Basic familiarity with terminal commands.

* * *

### **Step 1: Install Ollama**

First, install Ollama to run AI models locally.

**For macOS/Linux:**

```bash
curl -fsSL https://ollama.com/install.sh | sh  

```

**For Windows (Preview):**  
Download the installer from [Ollama.com](https://ollama.com) and run it.

Start the Ollama service:

```bash
ollama serve  

```

_(Keep this terminal running in the background.)_

* * *

### **Step 2: Download the DeepSeek Model**

Pull the DeepSeek model from Ollama’s library. For example, use `deepseek-coder` (adjust the version as needed):

```bash
ollama pull deepseek-coder:33b-instruct-q4_K_M  

```

_(Replace with your preferred variant, e.g., `6.7b` or `1.3b` for lighter models.)_

* * *

### **Step 3: Install Continue in VSCode**

1. Open VSCode.
2. Go to **Extensions** (Ctrl+Shift+X / Cmd+Shift+X).
3. Search for **“Continue”** and install it.

**Alternatively** , use the VSCode Command Palette (Ctrl+Shift+P / Cmd+Shift+P):

```bash
ext install continue  

```

* * *

### **Step 4: Configure Continue to Use Ollama & DeepSeek**

1. Open the Continue configuration file in VSCode:

2. Add Ollama as a model provider and specify DeepSeek:

```json
{  
  "models": [  
    {  
      "title": "DeepSeek via Ollama",  
      "provider": "ollama",  
      "model": "deepseek-coder:33b-instruct-q4_K_M",  
      "baseUrl": "http://localhost:11434"  
    }  
  ]  
}  

```

_(Match the `model` name to the one you downloaded in Step 2.)_

1. Save the file (`Ctrl+S` / `Cmd+S`).

* * *

### **Step 5: Test the Integration**

1. Create a test file (e.g., `test.py`).
2. Write a comment prompting DeepSeek:

```python
# Write a function to reverse a string in Python  

```

1. Place your cursor on the comment line and press `Ctrl+Shift+I` (or `Cmd+Shift+I` on macOS) to trigger Continue.
2. DeepSeek will generate code suggestions!

* * *

### **Troubleshooting Tips**

- **Ollama Isn’t Running** : Ensure `ollama serve` is running in the background.
- **Model Not Found** : Double-check the model name in `config.json` and ensure you’ve pulled it via `ollama pull`.
- **Slow Responses** : Use smaller DeepSeek variants (e.g., `1.3b`) or upgrade hardware.

* * *

By following these steps, you’ve transformed VSCode into an AI-powered IDE with **DeepSeek** and **Ollama**! 🚀

For more details, visit the [Continue Documentation](https://continue.dev/docs) or [Ollama’s Blog](https://ollama.com/blog/continue-code-assistant).

---

<div class="post-metadata">

**Author:** ![Aikon](https://onehack.st/user_avatar/onehack.st/aikon/32/26065_2.png) [@Aikon](https://onehack.st/u/Aikon)\
**Post date:** [January 29, 2025, 12:48pm UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/2 "2025-01-29T12:48:15Z")

</div>

This I guess is a memory issue right? ☹

![image](https://onehack.st/uploads/default/original/3X/8/a/8a9daac71176ed350b9e11c432402ab18a8cf965.png)

---

<div class="post-metadata">

**Author:** ![Pratik\_Patnaik](https://onehack.st/user_avatar/onehack.st/pratik_patnaik/32/91937_2.png) [@Pratik\_Patnaik](https://onehack.st/u/Pratik_Patnaik)\
**Post date:** [January 30, 2025, 5:03am UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/3 "2025-01-30T05:03:46Z")

</div>

Did you run ollame serve? It has to run in the background for the extension to work

---

<div class="post-metadata">

**Author:** ![Aikon](https://onehack.st/user_avatar/onehack.st/aikon/32/26065_2.png) [@Aikon](https://onehack.st/u/Aikon)\
**Post date:** [January 30, 2025, 6:31am UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/4 "2025-01-30T06:31:50Z")

</div>

yes it is…

 ![image](https://onehack.st/uploads/default/original/3X/8/2/82983edea4a408436eede36e5082aff01400e7e2.png)

and in fact it is being found. Unsure if I am doing something wrong…

 ![image](https://onehack.st/uploads/default/original/3X/6/2/62dc16b7b1df9eaba286ae941a1b1804669730ce.png)

---

<div class="post-metadata">

**Author:** ![tapan\_sharma](https://onehack.st/letter_avatar_proxy/v4/letter/t/73ab20/32.png) [@tapan\_sharma](https://onehack.st/u/tapan_sharma)\
**Post date:** [January 30, 2025, 1:09pm UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/5 "2025-01-30T13:09:44Z")

</div>

I figured it out, we also need to configure the tabAutocomplete Config:

```auto
"tabAutocompleteModel": {
    "title": "DeepSeek via Ollama",
    "provider": "ollama",
    "model": "deepseek-coder",
    "apiBase": "http://localhost:11434/"
  },

```

Here’s my config, for the 1.3b model, you can adjust it accordingly.

---

<div class="post-metadata">

**Author:** ![Aikon](https://onehack.st/user_avatar/onehack.st/aikon/32/26065_2.png) [@Aikon](https://onehack.st/u/Aikon)\
**Post date:** [January 30, 2025, 1:22pm UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/6 "2025-01-30T13:22:00Z")

</div>

still the same error … here is the top part of config.json.

 ![image](https://onehack.st/uploads/default/original/3X/5/2/5249c592b614cfb393805e83745d52b79c5578d1.png)

---

<div class="post-metadata">

**Author:** ![Aikon](https://onehack.st/user_avatar/onehack.st/aikon/32/26065_2.png) [@Aikon](https://onehack.st/u/Aikon)\
**Post date:** [January 30, 2025, 1:44pm UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/7 "2025-01-30T13:44:03Z")

</div>

I believe this is a memory issue. Will try a lighter model and get back with results.

---

<div class="post-metadata">

**Author:** ![Aikon](https://onehack.st/user_avatar/onehack.st/aikon/32/26065_2.png) [@Aikon](https://onehack.st/u/Aikon)\
**Post date:** [January 30, 2025, 2:13pm UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/8 "2025-01-30T14:13:19Z")

</div>

ok managed … used this one: 6.7b-base. It develops even in BASIC 😃 the times of the Commodore 64!!!

 ![image](https://onehack.st/uploads/default/original/3X/6/1/6155f09cea0e639b44b3a675a676af9ac2de12ac.png)

and removed the other one : ollama rm deepseek-coder:33b-instruct-q4\_K\_M

Thanks @Pratik_Patnaik and @tapan_sharma for helping!

---

<div class="post-metadata">

**Author:** ![Kittu](https://onehack.st/user_avatar/onehack.st/kittu/32/140185_2.png) [@Kittu](https://onehack.st/u/Kittu)\
**Post date:** [January 31, 2025, 4:22pm UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/9 "2025-01-31T16:22:17Z")

</div>

How much space it has taken ?

---

<div class="post-metadata">

**Author:** ![Aikon](https://onehack.st/user_avatar/onehack.st/aikon/32/26065_2.png) [@Aikon](https://onehack.st/u/Aikon)\
**Post date:** [February 2, 2025, 10:11am UTC](https://onehack.st/t/using-deepseek-locally-and-vscode-extension/299926/10 "2025-02-02T10:11:08Z")

</div>

the one installed last is nearly 4GB … the other one was much much more…

 ![image](https://onehack.st/uploads/default/original/3X/6/e/6e5b073c2a887b11be07ef5abcfabbe3c2ecc3b0.png)