Steve Chen's picture

4

Steve Chen

stev236

AI & ML interests

Running local models on different projects.

Recent Activity

new activity 17 days ago

Qwen/Qwen3-8B:New 8B model much slower than old 7B model when running on vLLM.

new activity 17 days ago

Qwen/Qwen3-4B:Why are the new 4B and 8B models slower than the previous 7B-1M model??

new activity 18 days ago

Qwen/Qwen3-14B:Long context: YaRN max_position_embeddings 32K or 40k?

View all activity

Organizations

None yet

stev236's activity

New activity in Qwen/Qwen3-8B 17 days ago

New 8B model much slower than old 7B model when running on vLLM.

#6 opened 17 days ago by

New activity in Qwen/Qwen3-4B 17 days ago

Why are the new 4B and 8B models slower than the previous 7B-1M model??

#6 opened 17 days ago by

New activity in Qwen/Qwen3-14B 18 days ago

Long context: YaRN max_position_embeddings 32K or 40k?

#10 opened 18 days ago by

New activity in mistralai/Mistral-Small-3.1-24B-Instruct-2503 about 1 month ago

vLLM example for 'Offline' should include an input image.

#47 opened about 2 months ago by

vLLM example for 'Offline' should include an input image.

#47 opened about 2 months ago by