opensource/kilocode
3
0
Fork 0
mirror of https://github.com/Kilo-Org/kilocode.git synced 2026-08-28 11:05:31 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'main' into flint-zenith

Browse Source
This commit is contained in:
Joshua Lambert
2026-04-27 19:19:25 -04:00
committed by GitHub
parent 4e96ed8ce9 f0bdc5a8e8
commit dda7ec8a6e
1 changed files with 21 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
packages/kilo-docs/pages/code-with-ai/agents/custom-models.md
+21
View File
@@ -84,6 +84,7 @@ All fields are optional. When a model ID matches one already in the built-in cat
| `reasoning` | `boolean` | Whether the model supports extended thinking |
| `temperature` | `boolean` | Whether the model supports the temperature parameter |
| `attachment` | `boolean` | Whether the model supports file attachments |
| `modalities` | `object` | Optional. Supported input and output types: `{ input, output }` |
| `limit` | `object` | Token limits: `{ context, output, input? }` |
| `cost` | `object` | Pricing per million tokens: `{ input, output, cache_read?, cache_write? }` |
| `options` | `object` | Arbitrary provider-specific model options |
@@ -91,6 +92,26 @@ All fields are optional. When a model ID matches one already in the built-in cat
| `provider` | `object` | Override `{ npm?, api? }` — the AI SDK package or base API URL for this model |
| `variants` | `object` | Named variant configurations (e.g., different reasoning efforts) |
### Modalities (modalities)
The `modalities` object declares which content types the model can receive and produce. It is optional — omit it to use defaults from the catalog or fallback to text-only. When `modalities` is provided, both `input` and `output` arrays are required. Each array can include `text`, `image`, `audio`, `video`, or `pdf`.
| Sub-field | Type | Required | Description |
| --------- | ------- | ---------------- | ------------------------------------------------ |
| `input` | `array` | Yes (if present) | Content types the model accepts from the user |
| `output` | `array` | Yes (if present) | Content types the model can generate in response |
For a standard text model that can also inspect images, use:
```jsonc
"modalities": {
"input": ["text", "image"],
"output": ["text"]
}
```
If `modalities` is omitted and the model ID matches a models.dev catalog entry for that provider, Kilo uses the catalog's modalities. For completely custom models with no catalog match, Kilo defaults to text input and text output only. Set `attachment: true` alongside image, audio, video, or PDF input modalities when the provider supports sending those files as attachments.
### Token Limits (limit)
The `limit` object controls how Kilo manages the model's context window and output length. These values are specified in **tokens**.