Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference
Paper • 2604.07394 • Published • 16
Flux Attention
🚀 Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference
-
-
QQTang1223/full_streaming_Llama-3.1-8B-Instruct
Text Generation • 8B • Updated • 296 -
QQTang1223/full_xattn_Qwen3-8B
Text Generation • 8B • Updated • 311 • 1 -
QQTang1223/full_xattn_Qwen3-4B
Text Generation • 4B • Updated • 293 -
QQTang1223/full_xattn_Llama-3.1-8B-Instruct
Text Generation • 8B • Updated • 296 -
QQTang1223/full_triangle_Qwen3-8B
Text Generation • 8B • Updated • 311 -
QQTang1223/full_triangle_Qwen3-4B
Text Generation • 4B • Updated • 330 -
QQTang1223/full_triangle_Llama-3.1-8B-Instruct
Text Generation • 8B • Updated • 330 -
QQTang1223/full_streaming_Qwen3-8B
Text Generation • 8B • Updated • 285 -
QQTang1223/full_streaming_Qwen3-4B
Text Generation • 4B • Updated • 332 -
QQTang1223/qwen_mix_sft_64K6
Viewer • Updated • 49.3k • 36 -
QQTang1223/llama_mix_sft_64K6
Viewer • Updated • 49.3k • 44 -
QQTang1223/xattn_streaming_Qwen3-4B
Text Generation • 4B • Updated • 246