meituan
/

DeepSeek-R1-Channel-INT8

Text Generation

Model card Files Files and versions Community

Resources

View closed (4)

Update inference/bf16_cast_channel_int8.py

#10 opened about 10 hours ago by

Update config.json

#9 opened about 10 hours ago by

how to achieve 2500 tps throughput？

#8 opened 1 day ago by

can this model run with `ollama` with `pure cpu` model?

#7 opened 4 days ago by

Add `quantization_config` in config.json?

#4 opened 7 days ago by

运行channel INT8后sglang报错OOM

#3 opened 9 days ago by