-
Notifications
You must be signed in to change notification settings - Fork 251
Issues: InternLM/lmdeploy
[Benchmark] benchmarks on different cuda architecture with mo...
#815
opened Dec 11, 2023 by
lvhan028
Open
6
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
How is the support for RoPE difference between
hf llama
and meta llama
?
#1700
opened May 31, 2024 by
sleepwalker2017
[Bug] ModuleNotFoundError: No module named '_turbomind' loading llava Mistral 7B
#1699
opened May 31, 2024 by
Alexis-IMBERT
2 tasks done
InternVL模型在推理时是否可以控制max_tiles参数等或者是否可以pipline直接传入pixel_values
#1696
opened May 31, 2024 by
YuMingtao0503
[Feature] InternVL-Chat-V1-5-AWQ merge LoRA adapter
awaiting response
#1691
opened May 30, 2024 by
isongxw
总是看到一个using default GEMM algo的WARNING,是否会因为使用了默认的GEMM而影响速度或者吞吐量?
#1680
opened May 29, 2024 by
lzcchl
[Bug] result of W4A16 quantized Qwen1.5-1.8B-Chat model not correct
#1676
opened May 29, 2024 by
dawnranger
2 tasks done
[Bug] lmdeploy lite auto_awq: RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0!
#1675
opened May 29, 2024 by
dawnranger
2 tasks done
[feature] need rope_scaling_factor args in benchmark/profile_generation.py to enable dynamic NTK.
#1670
opened May 28, 2024 by
zhulinJulia24
2 tasks
[Bug] WSL2环境下,0.4.2做InternVL量化时,磁盘写入速度极低
awaiting response
#1667
opened May 28, 2024 by
ysyx2008
2 tasks
Batch infer seems no speed up
awaiting response
#1665
opened May 28, 2024 by
xiangqi1997
1 of 2 tasks
LMDeploy-0.4.1运行qwen1.5 110B,推理长时间无结果
awaiting response
#1639
opened May 22, 2024 by
summerrain321
2 tasks done
Previous Next
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.