update readme

README.md

@@ -51,7 +51,6 @@ The model generates structured, schema-consistent JSON outputs for every video f
 - **Production-ready** - Battle-tested on trillion-scale video frame captioning workloads
 - **Schema-consistent JSON** - Reliable structured output for every frame
 - **Cost-efficient** - Optimized for high-throughput inference
-- **Temporal consistency** - Maintains semantic coherence across video sequences
 - **Open source** - Build and deploy without proprietary API dependencies
 
 ## Architecture
@@ -223,28 +222,15 @@ Given a nature scene with a wooden boardwalk through grassland:
 - **Video Analytics** - Extract insights from large video collections
 - **Content Management** - Automatic tagging and organization of video libraries
 
-## Limitations
+## Interested in training your own model?
 
-- Processes one video frame per request
-- English-only descriptions (can identify text in other languages)
-- Maximum image size: 1MB
-- Requires specific prompts for optimal performance
-- Not supported on A100 GPUs (no native FP8)
-
-## Best Practices
-
-1. **Use exact prompts** - The provided system and user prompts are optimized for best results
-2. **Set low temperature** - Use temperature=0.1 for consistent outputs
-3. **Enable JSON mode** - Always set response_format to ensure valid JSON
-4. **Process systematically** - Maintain temporal order when processing video sequences
-5. **Batch similar content** - Group frames from the same video for efficiency
+Contact us at [sales@inference.net](mailto:sales@inference.net) for a free consultation with our research team.
 
 ## Support
 
-- **Documentation**: [docs.inference.net](https://docs.inference.net)
-- **API Access**: [inference.net/use-cases/video-understanding](https://localhost:3000/use-cases/video-understanding)
+- **Documentation**: [docs.inference.net](https://inference.net/use-cases/video-understanding)
+- **API Access**: [inference.net/use-cases/video-understanding](https://inference.net/use-cases/video-understanding)
 - **Email**: support@inference.net
-- **Enterprise**: [Schedule a consultation](https://inference.net/sales)
 
 ## License