Li dongyang

lsm03624's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in deepreinforce-ai/Ornith-1.0-35B-FP8 3 days ago

FP8 seems to be broken

3

#1 opened 3 days ago by

Neiko2002

New activity in lovedheart/Qwen-AgentWorld-35B-A3B-FP8 4 days ago

You didn't quantize it, did you? FP8 can't be the same size as BF16.

1

#1 opened 4 days ago by

lsm03624

New activity in WeiboAI/VibeThinker-3B 11 days ago

怎么自我认知还是deepseek？而且好像没有做快慢思考，无法自适应控制思考长度

3

#12 opened 11 days ago by

user48271

New activity in cyankiwi/Qwen3-Coder-Next-AWQ-4bit 4 months ago

模型量化的效果并不理想

5

#2 opened 5 months ago by

mediali

New activity in cyankiwi/Qwen3.5-35B-A3B-AWQ-8bit 4 months ago

The VLLM installed in this way can run this model.

#1 opened 4 months ago by

lsm03624

New activity in cyankiwi/Qwen3.5-35B-A3B-AWQ-4bit 4 months ago

Unable to run (vllm/sglang)

8

#1 opened 4 months ago by

nfunctor

New activity in cyankiwi/Qwen3-Coder-Next-AWQ-4bit 5 months ago

Can we perform 4-bit quantization for the awq of the Step-3.5-Flash model? The VLLM can run it.

1

#3 opened 5 months ago by

lsm03624

New activity in ubergarm/Kimi-K2-Thinking-GGUF 8 months ago

the faster the inference speed becomes. Why is that?

👍 1

1

#9 opened 8 months ago by

lsm03624

New activity in MidnightPhreaker/GLM-4.5-Air-REAP-82B-A12B-GPTQ-INT4-gs32 8 months ago

GPU: 5060Ti*4, vllm version 0.11. The following error occurred:

1

#1 opened 8 months ago by

lsm03624

New activity in cyankiwi/Kimi-Linear-48B-A3B-Instruct-AWQ-4bit 8 months ago

Error installing from PR branch

👍 1

14

#1 opened 8 months ago by

DrRos

New activity in QuantTrio/GLM-4.5-Air-AWQ-FP16Mix 8 months ago

Thanks!

❤️ 2

9

#1 opened 11 months ago by

lightenup

New activity in DevQuasar/MiniMaxAI.MiniMax-M2-GGUF 8 months ago

Seems to be working with PR and `--jinja`

❤️🤝 3

4

#1 opened 8 months ago by

ubergarm