luchangli03's picture

luchangli03

luchangli03

·

AI & ML interests

None yet

Recent Activity

liked a model 2 months ago

lightseekorg/kimi-k2.5-eagle3

new activity 2 months ago

AQ-MedAI/Kimi-K25-eagle3:Can you reduce the kv head num of this model? "num_key_value_heads": 64, which requies a lots of kv cache

liked a model 2 months ago

AQ-MedAI/Kimi-K25-eagle3

View all activity

Organizations

liked a model 2 months ago

lightseekorg/kimi-k2.5-eagle3

3B • Updated Mar 16 • 60.3k • 11

New activity in AQ-MedAI/Kimi-K25-eagle3 2 months ago

Can you reduce the kv head num of this model? "num_key_value_heads": 64, which requies a lots of kv cache

#1 opened 2 months ago by

liked a model 2 months ago

AQ-MedAI/Kimi-K25-eagle3

1B • Updated Mar 30 • 12.1k • 10

New activity in jerryzh168/Kimi-K2-Thinking-FP8 3 months ago

Can you provide the code that convert the int4 weight to fp8? thanks

#2 opened 3 months ago by

New activity in chutesai/DeepSeek-V3.1-Terminus-NextN 5 months ago

what's the difference between this nextn and self contained MTP model in DeepSeek-V3.1-Terminus?

#1 opened 5 months ago by

liked a Space about 1 year ago

The Ultra-Scale Playbook

The ultimate guide to training LLM on large GPU Clusters