ltcai 0.1.27 → 0.1.28

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/docs/CHANGELOG.md CHANGED
@@ -1,5 +1,20 @@
1
1
  # Changelog
2
2
 
3
+ ## [0.1.28] - 2026-05-24
4
+
5
+ ### 버그 수정: 추천 모델 ID 오류
6
+
7
+ - **`google/gemma-4-E4B` → `mlx-community/gemma-4-e4b-it-4bit` 수정**
8
+ - 기존 ID는 HuggingFace의 BF16 풀프리시전 원본 모델 (~16GB) 로, MLX 포맷이 아니어서 `mlx_vlm.load()` 로 로드 불가능
9
+ - 올바른 MLX 4-bit 양자화 버전(`mlx-community/gemma-4-e4b-it-4bit`, 5.2GB, 43K downloads)으로 교체
10
+ - 크기 표시도 `"Next-Gen"` → `"5.2GB"` 로 실제 값으로 수정
11
+
12
+ ### Release
13
+ - 배포 버전을 `0.1.28`로 상향
14
+ - 대상 채널: `npm` · `PyPI` · `VS Code Marketplace` · `Open VSX`
15
+
16
+ ---
17
+
3
18
  ## [0.1.27] - 2026-05-24
4
19
 
5
20
  ### 로그인 페이지 UI 개선
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "ltcai",
3
- "version": "0.1.27",
3
+ "version": "0.1.28",
4
4
  "description": "Lattice AI local MLX/cloud LLM workspace server",
5
5
  "homepage": "https://github.com/TaeSooPark-PTS/LatticeAI#readme",
6
6
  "repository": {
package/server.py CHANGED
@@ -4000,7 +4000,7 @@ async def list_models():
4000
4000
  {"id": "mlx-community/Llama-3.1-8B-Instruct-4bit", "name": "Llama 3.1 8B", "tag": "general", "size": "4.7GB"},
4001
4001
 
4002
4002
  # Gemma Series
4003
- {"id": "google/gemma-4-E4B", "name": "Gemma 4 E4B (Latest)", "tag": "next-gen", "size": "Next-Gen"},
4003
+ {"id": "mlx-community/gemma-4-e4b-it-4bit", "name": "Gemma 4 E4B (4-bit)", "tag": "next-gen", "size": "5.2GB"},
4004
4004
  {"id": "mlx-community/gemma-2-9b-it-4bit", "name": "Gemma 2 9B", "tag": "balanced","size": "5.4GB"},
4005
4005
  {"id": "mlx-community/gemma-2-2b-it-4bit", "name": "Gemma 2 2B", "tag": "ultra-light", "size": "1.6GB"},
4006
4006