LMSYS-Chat
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Chatbot workload | LMSYS-Chat-1M | Average PTLA (s/token)0.47 | 28 | |
| Text Compression | LMSYS-Chat (Cluster 9: Casual Q&A) | Compression Ratio0.1 | 6 | |
| Text Compression | LMSYS-Chat Cluster 8: Translation Language | Compression Ratio0.08 | 6 | |
| Text Compression | LMSYS-Chat Cluster 7: Science/Math | Compression Ratio0.39 | 6 | |
| Text Compression | LMSYS-Chat Cluster 6: Philosophy/Ethics | Compression Ratio0.09 | 6 | |
| Text Compression | LMSYS-Chat (Cluster 5: Business/Professional) | Compression Ratio0.09 | 6 | |
| Text Compression | LMSYS-Chat Cluster 4: Roleplay Fiction | Compression Ratio0.4 | 6 | |
| Text Compression | LMSYS-Chat Cluster 3: Academic/Education | Compression Ratio0.09 | 6 | |
| Text Compression | LMSYS-Chat Cluster 2: Code/Technical | Compression Ratio0.1 | 6 | |
| Text Compression | LMSYS-Chat Cluster 1: Creative Writing | Compression Ratio0.11 | 6 | |
| Text Compression | LMSYS-Chat Cluster 0: General Chat | Compression Ratio0.11 | 6 | |
| Text Compression | LMSYS-Chat Overall | Compression Ratio0.09 | 6 | |
| LLM Inference Scheduling | LMSYS-Chat-1M | Average Per-token Latency (s/token)2.41 | 4 | |
| Toxicity Detection | LMSYS-Chat-1M | Accuracy0.9669 | 4 | |
| Complexity prediction | LMSYS-CHAT-1M | ROC-AUC90.1 | 3 |
Showing 15 of 15 rows