-
Notifications
You must be signed in to change notification settings - Fork 42
Pull requests: sgl-project/sgl-kernel-xpu
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
DeepSeek V4 Update plan_compress prefill and decode for compress ratio 128
#317
opened Jul 21, 2026 by
polisettyvarma
Contributor
Loading…
[MLA] Support sparse MLA for prefill and decode
#316
opened Jul 21, 2026 by
Valentine233
Collaborator
Loading…
[XPU] Add fused MXFP8 W8A16 GEMM (moe_grouped_mm_nt_xe20_mxfp8_w8a16)
#311
opened Jul 18, 2026 by
kalyank007
•
Draft
Support per token group quant 8bit v2 jit kernel
run-ci
#310
opened Jul 17, 2026 by
kareemshaik80
Collaborator
Loading…
Apply jit per_tensor_quant_fp8 kernel
run-ci
#309
opened Jul 17, 2026 by
kareemshaik80
Collaborator
Loading…
Replace chunk prefill with AppendKV FMHA prefill
run-ci
#307
opened Jul 17, 2026 by
yuankuns
Collaborator
Loading…
[MoE] fix wrong index when expert_num=16
run-ci
#295
opened Jul 15, 2026 by
airMeng
Collaborator
Loading…
Add jit support for per_token_group_quant_8bit
run-ci
#294
opened Jul 15, 2026 by
kareemshaik80
Collaborator
Loading…
[Internal/CRI] Changes to build/enable CRI - WIP
#292
opened Jul 15, 2026 by
mkumargarg
Collaborator
•
Draft
WIP support for Min p sampling kernel
#288
opened Jul 14, 2026 by
krishnanand-k
Collaborator
•
Draft
added support for top_p_renorm and top_k_top_p_sampling
run-ci
#282
opened Jul 10, 2026 by
pralay-das
Collaborator
Loading…
kvcacheio: add XPU scatter/gather KV cache transfer kernels
run-ci
#261
opened Jul 2, 2026 by
sspintel
Collaborator
Loading…
3 tasks done
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.