@capgo/capacitor-llm 8.1.5 → 8.1.7
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md
CHANGED
|
@@ -1,12 +1,28 @@
|
|
|
1
1
|
# @capgo/capacitor-llm
|
|
2
2
|
|
|
3
|
-
|
|
3
|
+
Run large language models on the device from your Capacitor app: Apple Intelligence on iOS, Gemini Nano or LiteRT-LM on Android, and Gemma web models in the browser. Private and free per request.
|
|
4
|
+
|
|
5
|
+
<a href="https://capgo.app/?ref=plugin_llm"><img src="https://capgo.app/readme-banner.svg?repo=Cap-go/capacitor-llm" alt="Capgo - Instant updates for Capacitor" /></a>
|
|
4
6
|
|
|
5
7
|
<div align="center">
|
|
6
|
-
<
|
|
7
|
-
<h2><a href="https://capgo.app/
|
|
8
|
+
<p><b>Capgo</b>: open-source live updates for Ionic and Capacitor apps. Ship OTA fixes and features instantly, without waiting for app store review.</p>
|
|
9
|
+
<h2><a href="https://capgo.app/register/?ref=plugin_llm">➡️ Get started for free</a></h2>
|
|
10
|
+
<p>14-day unlimited free trial. No credit card required</p>
|
|
11
|
+
<p><a href="https://capgo.app/consulting/?ref=plugin_llm">Missing a feature? We'll build the plugin for you 💪</a></p>
|
|
8
12
|
</div>
|
|
9
13
|
|
|
14
|
+
<p align="center">
|
|
15
|
+
<img src="https://raw.githubusercontent.com/Cap-go/capacitor-llm/main/assets/github-social-preview.png" alt="@capgo/capacitor-llm for Capacitor apps" width="300" />
|
|
16
|
+
</p>
|
|
17
|
+
|
|
18
|
+
## Key features
|
|
19
|
+
|
|
20
|
+
- **Chat sessions**: `createChat()` and `sendMessage()`, with streamed `textFromAi`, `aiFinished` and `generationError` events.
|
|
21
|
+
- **Model choice**: `setModel()` picks Apple Intelligence, Gemini Nano or a custom `.litertlm` model.
|
|
22
|
+
- **Model download**: `downloadModel()` fetches a model to the device with `downloadProgress` events.
|
|
23
|
+
- **Readiness**: `getReadiness()` and the `readinessChange` event tell you when the model can answer.
|
|
24
|
+
- **Platforms**: iOS, Android and Web. Web uses Gemma models through `@mediapipe/tasks-genai`. It loads the runtime from a CDN and needs the model file, so it works offline only once both are available locally.
|
|
25
|
+
|
|
10
26
|
On-device LLM support for Capacitor.
|
|
11
27
|
|
|
12
28
|
Current platform strategy:
|
|
@@ -17,7 +17,7 @@ import MediaPipeTasksGenAI
|
|
|
17
17
|
|
|
18
18
|
@objc(LLMPlugin)
|
|
19
19
|
public class LLMPlugin: CAPPlugin, CAPBridgedPlugin {
|
|
20
|
-
private let pluginVersion: String = "8.1.
|
|
20
|
+
private let pluginVersion: String = "8.1.7"
|
|
21
21
|
private let liteRtTopP: Float = 0.95
|
|
22
22
|
|
|
23
23
|
public let identifier = "LLMPlugin"
|