Upload README.md with huggingface_hub

README.md

@@ -1,76 +1,48 @@
 ---
 language:
   - en
-license: mit
+license: apache-2.0
 library_name: onnx
 tags:
   - onnx
-  - summarization
-  - transformers.js
-  - in-browser
-  - quantized
-  - int8
-  - bart
   - affectively
-  - edgework
 base_model: facebook/bart-large-cnn
 pipeline_tag: summarization
-datasets:
-  - cnn_dailymail
 ---
 
-# BART Large CNN (ONNX, int8 Quantized)
+# Bart Large Cnn
 
-> **Production-ready** ONNX conversion of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) for in-browser text summarization — zero server cost, zero latency, complete privacy.
+AFFECTIVELY conversion of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) to ONNX format for edge deployment.
 
-## Highlights
+## Model Details
 
-- **Abstractive summarization** — generates concise summaries, not just extractive snippets
-- **~493 MB** quantized — BART-large architecture fine-tuned on CNN/DailyMail
-- **transformers.js compatible** — drop-in `pipeline('summarization')`
-- **406M parameters** — high-quality summaries from long-form text
+- **Source Model**: [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn)
+- **Format**: ONNX
+- **Converted by**: [AFFECTIVELY](https://affectively.ai)
 
-## Quick Start
+## Usage
 
-```javascript
-import { pipeline } from '@huggingface/transformers';
+### With ONNX Runtime
 
-const summarizer = await pipeline(
-  'summarization',
-  'affectively-ai/bart-large-cnn-onnx',
-  { dtype: 'q8' }
-);
+```python
+import onnxruntime as ort
 
-const result = await summarizer(
-  'Today was an incredibly challenging day at work. I had back-to-back meetings...',
-  { max_length: 50, min_length: 20 }
-);
-// [{ summary_text: 'The author had a challenging day with back-to-back meetings...' }]
+session = ort.InferenceSession("bart-large-cnn-onnx.onnx")
 ```
 
-## Conversion Details
+### With Transformers + Optimum
 
-| Property | Value |
-|----------|-------|
-| Base model | [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) |
-| Training data | CNN/DailyMail (300k news articles) |
-| Export | PyTorch → ONNX via [Optimum](https://huggingface.co/docs/optimum) |
-| Quantization | int8 dynamic (`ORTQuantizer`, avx512_vnni) |
-| Quantized size | ~493 MB |
-
-## Use Cases
-
-This model powers text summarization in [Edgework.ai](https://edgework.ai) — bringing fast, cheap, and private inference as close to the user as possible. Best for:
+```python
+from optimum.onnxruntime import ORTModelForSequenceClassification
+model = ORTModelForSequenceClassification.from_pretrained("affectively-ai/bart-large-cnn-onnx")
+```
 
-- Summarizing journal entries for daily/weekly emotion reports
-- Condensing therapy session notes
-- Creating digests from long conversation threads
-- Generating TL;DR for mental wellness content
+## About AFFECTIVELY
 
-## About
+[AFFECTIVELY](https://affectively.ai) is an emotion intelligence platform that runs AI models at the edge -- in-browser, on-device, zero cloud cost. These converted models power the platform's real-time emotion analysis, speech recognition, and natural language capabilities.
 
-Published by [AFFECTIVELY](https://huggingface.co/affectively-ai) · Managed by [@buley](https://huggingface.co/buley)
+All conversions are optimized for edge deployment within browser and mobile memory constraints.
 
-We convert, quantize, and publish **production-ready ONNX models** for edge and in-browser inference. Every release is tested for correctness and stability before publication.
+## License
 
-- [All models](https://huggingface.co/affectively-ai) · [GitHub](https://github.com/affectively-ai) · [Edgework.ai](https://edgework.ai)
+Apache 2.0 (follows upstream model license)