deepseek-ai/DeepGEMM
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepGEMM.git synced 2026-11-05 00:47:26 +00:00
Code Issues 162 Projects Packages Wiki Activity

[PR #200] [MERGED] Add DeepSeek V3.2 indexer logit kernels for H800 (SM90) #290

New issue
Closed
opened 2026-10-26 21:15:52 +00:00 by bsakaguchi · 0 comments
bsakaguchi commented 2026-10-26 21:15:52 +00:00
Owner
Copy link

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/DeepGEMM/pull/200
Author: @LyricZhao
Created: 9/29/2025
Status: ✅ Merged
Merged: 9/29/2025
Merged by: @LyricZhao

Base: main ← Head: sm90-v32

📝 Commits (1)

📊 Changes

10 files changed (+1602 additions, -0 deletions)

View changed files

📝 csrc/apis/attention.hpp (+163 -0)
📝 csrc/indexing/main.cu (+5 -0)
➕ csrc/jit_kernels/impls/smxx_clean_logits.hpp (+79 -0)
➕ csrc/jit_kernels/impls/smxx_fp8_mqa_logits.hpp (+152 -0)
➕ csrc/jit_kernels/impls/smxx_fp8_paged_mqa_logits.hpp (+228 -0)
📝 deep_gemm/__init__.py (+4 -0)
➕ deep_gemm/include/deep_gemm/impls/sm90_fp8_mqa_logits.cuh (+317 -0)
➕ deep_gemm/include/deep_gemm/impls/sm90_fp8_paged_mqa_logits.cuh (+403 -0)
➕ deep_gemm/include/deep_gemm/impls/smxx_clean_logits.cuh (+67 -0)
📝 tests/test_attention.py (+184 -0)

📄 Description

No description provided

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/deepseek-ai/DeepGEMM/pull/200 **Author:** [@LyricZhao](https://github.com/LyricZhao) **Created:** 9/29/2025 **Status:** ✅ Merged **Merged:** 9/29/2025 **Merged by:** [@LyricZhao](https://github.com/LyricZhao) **Base:** `main` ← **Head:** `sm90-v32` --- ### 📝 Commits (1) - [`7c95b14`](https://github.com/deepseek-ai/DeepGEMM/commit/7c95b14aa4a66edd7b682e5acdde62351ca81197) Add SM90 kernels ### 📊 Changes **10 files changed** (+1602 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `csrc/apis/attention.hpp` (+163 -0) 📝 `csrc/indexing/main.cu` (+5 -0) ➕ `csrc/jit_kernels/impls/smxx_clean_logits.hpp` (+79 -0) ➕ `csrc/jit_kernels/impls/smxx_fp8_mqa_logits.hpp` (+152 -0) ➕ `csrc/jit_kernels/impls/smxx_fp8_paged_mqa_logits.hpp` (+228 -0) 📝 `deep_gemm/__init__.py` (+4 -0) ➕ `deep_gemm/include/deep_gemm/impls/sm90_fp8_mqa_logits.cuh` (+317 -0) ➕ `deep_gemm/include/deep_gemm/impls/sm90_fp8_paged_mqa_logits.cuh` (+403 -0) ➕ `deep_gemm/include/deep_gemm/impls/smxx_clean_logits.cuh` (+67 -0) 📝 `tests/test_attention.py` (+184 -0) </details> ### 📄 Description _No description provided_ --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
bsakaguchi 2026-10-26 21:15:52 +00:00
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
public-sync-20261014
0930-release
nv_dev
fix-mega-moe-fence
public-release-2609
public-release-2607
public-release-2606
mega-update
public-release-260416
ko3n1g/ci/build-wheels
nv_dev_f8e8fb5
nv_dev_950d31a
nv_dev_67fc648
nv_dev_c491439
nv_dev_4ff3f54
v2.1.1.post3
v2.1.1.post2
v2.1.1.post1
v2.1.1
v2.1.0
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/DeepGEMM#290
No description provided.