update read

README.md

@@ -1,20 +1,65 @@
-README.md:
 ---
-title: FF Embeddings
-emoji: 🎲
-colorFrom: purple
-colorTo: blue
-sdk: gradio
-sdk_version: 4.44.0
-app_file: app.py
-pinned: false
+license: apache-2.0
+language:
+- en
+- zh
+- es
+- fr
+- de
+- ja
+- ko
+- ru
+- ar
+metrics:
+- brier_score
+base_model:
+- Qwen/Qwen3-Embedding-0.6B
+new_version: Qwen/Qwen3-Embedding-0.6B
+pipeline_tag: sentence-similarity
+library_name: sentence-transformers
+tags:
+- code
+- semantic-search
+- multilingual
+- ttrpg
+- classification
+- embeddings
 ---
 
-# FairFate Embeddings API
+# 🎲 FairFate Embeddings
 
-Semantic embeddings for tabletop RPG product classification.
+**Powered by Qwen3-Embedding-0.6B** 
 
-## Model
-- **Model:** sentence-transformers/all-MiniLM-L6-v2
-- **Dimensions:** 384
-- **Max Tokens:** 256
+High-performance semantic embeddings service for tabletop RPG products
+
+## 🌟 Highlights
+
+- 🏆 **#1 on MTEB** Multilingual Leaderboard (70.58 score)
+- 🌍 **100+ Languages** - English, Spanish, French, German, Chinese, Japanese, Korean, Arabic, and more
+- 📐 **1024 Dimensions** - Flexible output (32-1024)
+- 📚 **32K Context** - Massive text support (vs 512 for most models)
+- ⚡ **Instruction-Aware** - Customizable for RPG domain (1-5% accuracy boost)
+- 🎯 **Domain-Optimized** - Pre-configured for tabletop RPG content
+
+## 🎯 Perfect For
+
+- **Product Classification**: Auto-tag RPG products by genre, game system, themes
+- **Semantic Search**: Find products by meaning, not just keywords
+- **Recommendations**: "Products similar to this" based on content
+- **Duplicate Detection**: Find similar or duplicate listings
+- **Content Moderation**: Detect inappropriate content
+- **Multilingual Matching**: Cross-language product similarity
+
+## 📊 Model Information
+
+| Property | Value |
+|----------|-------|
+| **Model** | [`Qwen/Qwen3-Embedding-0.6B`](https://huggingface.co/Qwen/Qwen3-Embedding-0.6B) |
+| **Dimensions** | 1024 (flexible: 32-1024) |
+| **Max Tokens** | 32,768 |
+| **Languages** | 100+ |
+| **Parameters** | 0.6B |
+| **Layers** | 28 |
+| **Normalization** | L2 normalized |
+| **Distance Metric** | Cosine similarity |
+| **MTEB Score** | 70.58 (multilingual) |