Skip to content

Commit 2122bf8

Browse files
fpagnyfpagny
authored
feat(genapi): update continue tutorial with latest models (#5354)
* feat(genapi): update continue tutorial with latest models and configuration * fix(genapi): add devstral in json snippets * fix(genapi): add openai as continue provider plugin
1 parent ee6b0db commit 2122bf8

File tree

1 file changed

+103
-41
lines changed

1 file changed

+103
-41
lines changed

‎pages/generative-apis/reference-content/adding-ai-to-vscode-using-continue.mdx‎

Lines changed: 103 additions & 41 deletions
Original file line numberDiff line numberDiff line change
@@ -36,56 +36,123 @@ To link Continue with Scaleway's Generative APIs, you can use built-in menus fro
3636
- Click **Continue** in the menu on the left.
3737
- In the prompt section, click on **Select model** dropdown, then on **Add Chat model**.
3838
- Select **Scaleway** as provider.
39-
- Select the model you want to use (we recommend `Qwen 2.5 Coder 32b` to get started with).
39+
- Select the model you want to use (we recommend `Qwen 2.5 Coder 32b` to get started with chat and autocompletion only).
4040
- Enter your **Scaleway secret key**.
4141
<Message type="tip">
4242
To start with, we recommend you use a Scaleway secret key having access to your `default` Scaleway project.
4343
</Message>
4444

45-
These actions will edit automatically your `config.json` file. To edit it manually, see [Configure Continue through configuration file](#configure-continue-through-configuration-file).
45+
These actions will automatically edit your `config.yaml` file. To edit it manually, see [Configure Continue through configuration file](#configure-continue-through-configuration-file).
4646

4747
<Message type="note">
48-
Embeddings and autocomplete models are not yet supported through graphical interface configuration. To enable them, you need to edit the configuration manually, see [Configure Continue through configuration file](#configure-continue-through-configuration-file).
48+
Agents, embeddings, and autocomplete models are not yet supported through graphical interface configuration. Manually edit the configuration to enable them. See [Configure Continue through configuration file](#configure-continue-through-configuration-file) for more information.
4949
</Message>
5050

5151
#### Configure Continue through a configuration file
5252

5353
To link Continue with Scaleway's Generative APIs, you can configure a settings file:
5454

55-
- Create a `config.json` file inside your `.continue` directory.
56-
- Add the following configuration to enable Scaleway's Generative API:
57-
```json
58-
{
59-
"models": [
60-
{
61-
"model": "qwen2.5-coder-32b-instruct",
62-
"title": "Qwen2.5 Coder",
63-
"provider": "scaleway",
64-
"apiKey": "###SCW_SECRET_KEY###"
65-
}
66-
],
67-
"embeddingsProvider": {
68-
"model": "bge-multilingual-gemma2",
69-
"provider": "scaleway",
70-
"apiKey": "###SCW_SECRET_KEY###"
71-
},
72-
"tabAutocompleteModel": {
73-
"model": "qwen2.5-coder-32b",
74-
"title": "Qwen2.5 Coder Autocomplete",
75-
"provider": "scaleway",
76-
"apiKey": "###SCW_SECRET_KEY###"
77-
}
78-
}
55+
- Open your `config.yaml` settings file:
56+
- If you have already configured a **Local Assistant**, click **Local Assistant**, then click the **wheel icon** to open your existing `config.yaml`
57+
- Otherwise, create a `config.yaml` file inside your `.continue` directory.
58+
- Add the following configuration to enable Scaleway's Generative API. This configuration uses three different models for each tasks:
59+
- `devstral-small-2505` for agentic workflows through a chat interface
60+
- `qwen2.5-coder-32b` for autocompletion when editing a file
61+
- `bge-multilingual-gemma2` for embedding and retrieving code context
62+
```yaml
63+
name: Continue Config
64+
version: 0.0.1
65+
models:
66+
- name: Devstral - Scaleway
67+
provider: openai
68+
model: devstral-small-2505
69+
apiBase: https://api.scaleway.ai/v1/
70+
apiKey: ###SCW_SECRET_KEY###
71+
defaultCompletionOptions:
72+
maxTokens: 8000
73+
contextLength: 50000
74+
roles:
75+
- chat
76+
- apply
77+
- embed
78+
- edit
79+
capabilities:
80+
- tool_use
81+
- name: Autocomplete - Scaleway
82+
provider: openai
83+
model: qwen2.5-coder-32b
84+
apiBase: https://api.scaleway.ai/v1/
85+
apiKey: ###SCW_SECRET_KEY###
86+
defaultCompletionOptions:
87+
maxTokens: 8000
88+
contextLength: 50000
89+
roles:
90+
- autocomplete
91+
- name: Embeddings Model - Scaleway
92+
provider: openai
93+
model: bge-multilingual-gemma2
94+
apiBase: https://api.scaleway.ai/v1/
95+
apiKey: ###SCW_SECRET_KEY###
96+
roles:
97+
- embed
98+
embedOptions:
99+
maxChunkSize: 256
100+
maxBatchSize: 32
101+
context:
102+
- provider: problems
103+
- provider: tree
104+
- provider: url
105+
- provider: search
106+
- provider: folder
107+
- provider: codebase
108+
- provider: web
109+
params:
110+
n: 3
111+
- provider: open
112+
params:
113+
onlyPinned: true
114+
- provider: docs
115+
- provider: terminal
116+
- provider: code
117+
- provider: diff
118+
- provider: currentFile
79119
```
80120
- Save the file at the correct location:
81-
- Linux/macOS: `~/.continue/config.json`
82-
- Windows: `%USERPROFILE%\.continue\config.json`
121+
- Linux/macOS: `~/.continue/config.yaml`
122+
- Windows: `%USERPROFILE%\.continue\config.yaml`
123+
- In **Local Assistant**, click on **Reload config** or restart VS Code.
124+
125+
Alternatively, a `config.json` file can be used with the following format. Note that this format is deprecated, and we recommend using `config.yaml` instead.
126+
```json
127+
{
128+
"models": [
129+
{
130+
"model": "devstral-small-2505",
131+
"title": "Devstral - Scaleway",
132+
"provider": "openai",
133+
"apiKey": "###SCW_SECRET_KEY###"
134+
}
135+
],
136+
"embeddingsProvider": {
137+
"model": "bge-multilingual-gemma2",
138+
"provider": "openai",
139+
"apiKey": "###SCW_SECRET_KEY###"
140+
},
141+
"tabAutocompleteModel": {
142+
"model": "qwen2.5-coder-32b",
143+
"title": "Autocomplete - Scaleway",
144+
"provider": "openai",
145+
"apiKey": "###SCW_SECRET_KEY###"
146+
}
147+
}
148+
```
83149

84150
<Message type="tip">
85-
For more details on configuring `config.json`, refer to the [official Continue documentation](https://docs.continue.dev/reference).
151+
For more details on configuring `config.yaml`, refer to the [official Continue documentation](https://docs.continue.dev/reference).
86152
If you want to limit access to a specific Scaleway Project, you should add the field `"apiBase": "https://api.scaleway.ai/###PROJECT_ID###/v1/"` for each model (ie. `models`, `embeddingsProvider` and `tabAutocompleteModel`) since the default URL `https://api.scaleway.ai/v1/` can only be used with the `default` project.
87153
</Message>
88154

155+
89156
### Activate Continue in VS Code
90157

91158
After configuring the API, open VS Code and activate Continue:
@@ -99,15 +166,10 @@ After configuring the API, open VS Code and activate Continue:
99166

100167
### Going further
101168

102-
You can add additional parameters to configure your model behaviour by editing `config.json`.
103-
For instance, you can add the following `systemMessage` value to modify LLM messages `"role":"system"` and/or `"role":"developer"` and provide less verbose answers:
104-
```json
105-
{
106-
"models": [
107-
{
108-
"model": "...",
109-
"systemMessage": "You are an expert software developer. You give concise responses."
110-
}
111-
]
112-
}
169+
You can add more parameters to configure your model's behavior by editing `config.yaml`.
170+
For instance, you can add the following `chatOptions.baseSystemMessage` value to modify LLM messages `"role":"system"` and/or `"role":"developer"` and provide less verbose answers:
171+
```yaml
172+
model:...
173+
chatOptions:
174+
baseSystemMessage: "You are an expert developer. Only write concise answers."
113175
```

0 commit comments

Comments
 (0)