Document Gallery 1.0.16 direct Hugging Face import + desktop LiteRT-LM CLI (serve/run)
Browse files
README.md
CHANGED
|
@@ -61,6 +61,8 @@ and ask a question.
|
|
| 61 |
|
| 62 |
## Run on Android — Google AI Edge Gallery
|
| 63 |
|
|
|
|
|
|
|
| 64 |
Run this model **with image input** in the official
|
| 65 |
[Google AI Edge Gallery](https://github.com/google-ai-edge/gallery) app — no custom app needed
|
| 66 |
(the bundle carries the tokenizer, chat template, and image preprocessing config):
|
|
@@ -73,6 +75,19 @@ Run this model **with image input** in the official
|
|
| 73 |
|
| 74 |
> **Tip:** on the **GPU** backend use one image per conversation (a known GPU-delegate trait of `fast_vlm` models); pick **CPU** if you want multiple images in one chat.
|
| 75 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
## Conversion notes
|
| 77 |
|
| 78 |
- LiteRT-LM `fast_vlm` bundle: VISION_ENCODER (`[1,448,448,3]`→`[1,256,4096]`) + VISION_ADAPTER
|
|
|
|
| 61 |
|
| 62 |
## Run on Android — Google AI Edge Gallery
|
| 63 |
|
| 64 |
+
> **Update (July 2026):** [Google AI Edge Gallery](https://github.com/google-ai-edge/gallery) **v1.0.16+** can import litert-lm models **directly from Hugging Face** inside the app (tap **+**) — no computer or `adb` needed. The manual steps below are only required on older builds or for sideloading a local file.
|
| 65 |
+
|
| 66 |
Run this model **with image input** in the official
|
| 67 |
[Google AI Edge Gallery](https://github.com/google-ai-edge/gallery) app — no custom app needed
|
| 68 |
(the bundle carries the tokenizer, chat template, and image preprocessing config):
|
|
|
|
| 75 |
|
| 76 |
> **Tip:** on the **GPU** backend use one image per conversation (a known GPU-delegate trait of `fast_vlm` models); pick **CPU** if you want multiple images in one chat.
|
| 77 |
|
| 78 |
+
## Run on desktop (LiteRT-LM CLI)
|
| 79 |
+
|
| 80 |
+
The same `.litertlm` bundle runs on macOS / Linux / Windows with the official
|
| 81 |
+
[LiteRT-LM CLI](https://github.com/google-ai-edge/LiteRT-LM) — including as a
|
| 82 |
+
local **OpenAI-compatible API server**:
|
| 83 |
+
|
| 84 |
+
```bash
|
| 85 |
+
pip install litert-lm
|
| 86 |
+
litert-lm import --from-huggingface-repo litert-community/InternVL3-1B InternVL3-1B.litertlm internvl3-1b
|
| 87 |
+
litert-lm run internvl3-1b # interactive chat in the terminal
|
| 88 |
+
litert-lm serve # local OpenAI-compatible API server
|
| 89 |
+
```
|
| 90 |
+
|
| 91 |
## Conversion notes
|
| 92 |
|
| 93 |
- LiteRT-LM `fast_vlm` bundle: VISION_ENCODER (`[1,448,448,3]`→`[1,256,4096]`) + VISION_ADAPTER
|