Skip to content

feat: add LightON OCR-2 1B model entries to registry #765

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
olyasir merged 4 commits into from
Mar 10, 2026
Merged

feat: add LightON OCR-2 1B model entries to registry #765

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
b7a15d7
feat: add LightON OCR-2 1B model entries to registry
olyasir Mar 9, 2026
35bbb75
Merge branch 'main' into feat/lighton-ocr2-registry
olyasir Mar 10, 2026
1e0b8c1
fix: correct JSON syntax in models.prod.json
olyasir Mar 10, 2026
c5b5d7d
Merge branch 'main' into feat/lighton-ocr2-registry
olyasir Mar 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
32 changes: 32 additions & 0 deletions packages/qvac-lib-registry-server/data/models.prod.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5339,6 +5339,38 @@
],
"notes": "Full INT8 quantization of https://huggingface.co/istupakov/parakeet-tdt-0.6b-v3-onnx"
},
{
"source": "https://huggingface.co/noctrex/LightOnOCR-2-1B-ocr-soup-GGUF/resolve/main/LightOnOCR-2-1B-ocr-soup-Q4_K_M.gguf",

@yuranich yuranich Mar 18, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please point to a specific commit, not just the main branch

"engine": "@qvac/llm-llamacpp",
"description": "OCR-specialized vision-language model for document understanding",
"quantization": "q4_k_m",
"params": "0.6B",
"license": "Apache-2.0",
"tags": [
"generation",
"multimodal",
"ocr",
"lighton-ocr-2"
],
"notes": "Qwen3 LLM (0.6B) + SigLIP vision encoder. Full-page OCR, 11 languages. Use --jinja for chat template.",
"link": "https://huggingface.co/noctrex/LightOnOCR-2-1B-ocr-soup-GGUF"
},
{
"source": "https://huggingface.co/noctrex/LightOnOCR-2-1B-ocr-soup-GGUF/resolve/main/mmproj-F16.gguf",
"engine": "@qvac/llm-llamacpp",
"description": "Vision projector for LightON OCR-2 1B",
"quantization": "f16",
"params": "0.6B",
"license": "Apache-2.0",
"tags": [
"generation",
"multimodal",
"ocr",
"lighton-ocr-2"
],
"notes": "F16 recommended — half the size of F32, negligible quality difference",
"link": "https://huggingface.co/noctrex/LightOnOCR-2-1B-ocr-soup-GGUF"
},
{
"source": "https://huggingface.co/mradermacher/AfriqueGemma-4B-GGUF/blob/e1324d25db75de68b604093e527272f6c5aba69f/AfriqueGemma-4B.Q4_K_M.gguf",
"engine": "@qvac/llm-llamacpp",
Expand Down
Loading