Skip to content

Issues: deepseek-ai/DeepEP

Easier Potential Overall Design
#39 opened Mar 4, 2025 by LyricZhao
Open
Beta
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.

Sign up for GitHub

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

55 Open 78 Closed
55 Open 78 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

illegal memory on low_latency_dispatch when test dataset
#152 opened May 10, 2025 by hpz4311
1
[Feature] When will Ampere GPUs be supported? Is there a specific roadmap?
#149 opened May 8, 2025 by Iamleos
2
low latency mode bandwidth
#148 opened May 8, 2025 by duzeyan
2
How does DeepEP ensure memory consistency
#146 opened May 5, 2025 by MaoZiming
1
Caught signal 11 (Segmentation fault: address not mapped to object at address 0x8000008f)
#145 opened Apr 30, 2025 by LG-RJ
1
Profiling Results on H20 with DSv3
#144 opened Apr 29, 2025 by hpz4311
1
when i installed nvshmem_src, nccl get error, how to fix?
#143 opened Apr 29, 2025 by MarsMeng1994
2
The first call to dispatch after a short period has high latency.
#138 opened Apr 27, 2025 by TheRainstorm
10
[test_internode.py] failed on multi-QP: dispatch timeout on ROCE network with testing 2*H20 nodes
#137 opened Apr 27, 2025 by jeffye-dev
22
L1::no_allocate is not safe to load volatile data
#136 opened Apr 27, 2025 by Yinghan-Li
3
Assertion failed: /DeepEP/csrc/kernels/internode.cu:483, condition: ibgda_get_state()->num_rc_per_pe >= num_channels
#134 opened Apr 26, 2025 by whybeyoung
2
3
error: nvshmem API called before nvshmem_init
#133 opened Apr 24, 2025 by guodongrui-gdr
5
Ways to do overlapping of computation and commucation.
#131 opened Apr 22, 2025 by kuozhang
1
[intranode] when num_tokens change to 65536, then check failed
#127 opened Apr 17, 2025 by elevenxiang
2
test_inter_node.py failed: an illegal memory access was encountered
#126 opened Apr 16, 2025 by josephydu
4
Why is there no fp8 combine?
#123 opened Apr 15, 2025 by thefacetakt
2
Crash when test DeepEP over 16 H100 Servers
#114 opened Apr 9, 2025 by yanminjia
1
some questions in ibgda
#113 opened Apr 8, 2025 by Thunderbrook
2
how can I run the test across multi nodes？
#106 opened Apr 2, 2025 by idealboy
4
deepEP run two nodes, each nodes equipped 1 A100+1CX6, hit assert.
#104 opened Mar 31, 2025 by Knight-Cai
4
python test_low_latency.py blocked in H20
#103 opened Mar 31, 2025 by cll24
1
Why normal mode does not support AR?
#100 opened Mar 31, 2025 by yejunhong1
4
Suggestions for Improving the Readability of DeepEP Code
#98 opened Mar 28, 2025 by zuochunwei
9
Question about the global synchronization/barrier in internode notify_dispatch
#94 opened Mar 26, 2025 by songhexiang
5
Will the Normal Kernel (dispatch kernel) exclusively occupy the SM of the GPU?
#92 opened Mar 25, 2025 by PhilipJohnson2
1
ProTip! Follow long discussions with comments:>50.