Long-Context Model Laboratory

community

https://github.com/LCM-Lab

Activity Feed Request to join this org

AI & ML interests

Long-context Modeling, Reinforcement-Learning, Multi-modality

Recent Activity

QQTang1223 authored a paper about 21 hours ago

Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference

QQTang1223 submitted a paper 1 day ago

Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference

QQTang1223 updated a collection 2 months ago

Elastic-Attention

View all activity

Papers

Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference

View all Papers

LCM-Lab 's models 20

LCM-Lab/full_streaming_64k_qwen3-4b_MLP3.0_wfrozen

Text Generation • 4B • Updated Jan 28 • 1

LCM-Lab/full_streaming_64k_qwen3-4b_MLP8.0_wfrozen

Text Generation • 4B • Updated Jan 28 • 2

LCM-Lab/moba_llama

Text Generation • Updated Jan 28 • 1

LCM-Lab/infllm_qwen3-8b

Text Generation • 8B • Updated Jan 28 • 1

LCM-Lab/moba_qwen3-8b

Text Generation • 8B • Updated Jan 28 • 4

LCM-Lab/infllm_llama

Text Generation • 8B • Updated Jan 28 • 1

LCM-Lab/infllm_qwen3-4b

Text Generation • 4B • Updated Jan 28 • 5

LCM-Lab/moba_qwen3-4b

Text Generation • 4B • Updated Jan 28 • 5

LCM-Lab/nsa_qwen3-8b

Text Generation • 9B • Updated Jan 28 • 1

LCM-Lab/nsa_qwen3-4b

Text Generation • Updated Jan 28

LCM-Lab/nsa_llama

Text Generation • 8B • Updated Jan 28 • 1

LCM-Lab/1.2steps300_full_streaming_64k_qwen3-8b_wfrozen

Text Generation • 8B • Updated Jan 28 • 1

LCM-Lab/full_streaming_64k_qwen3-4b_MLP2.0_wfrozen

Text Generation • 4B • Updated Jan 28 • 2

LCM-Lab/full_xattn_64k_qwen3-8b_end0.7_wfrozen

Text Generation • 8B • Updated Jan 28 • 2

LCM-Lab/full_streaming_64k_qwen3-4b_end0.7_wfrozen

Text Generation • 4B • Updated Jan 28 • 1

LCM-Lab/full_xattn_64k_qwen3-4b_wfrozen

Text Generation • 4B • Updated Jan 28 • 2

LCM-Lab/full_xattn_64k_llama3.1-8b_wfrozen

Text Generation • 8B • Updated Jan 28 • 2

LCM-Lab/LongReward_Skywork-Reward-V2-Llama-3.1-8B

8B • Updated Oct 11, 2025 • 1

LCM-Lab/LongReward_Con-J-Qwen2-7B

Text Generation • 8B • Updated Oct 11, 2025 • 4

LCM-Lab/LongReward_Qwen3-8B

Text Generation • 308k • Updated Oct 11, 2025 • 6 • 1