Skip to content

Pull requests: mirage-project/mirage

New pull request New
27 Open 348 Closed
27 Open 348 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
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Dynamically Sharded Model Weight Loader
#632 opened Feb 26, 2026 by EH67 • Draft
EP MoE
#631 opened Feb 25, 2026 by STWMichael • Draft
@STWMichael
Refactor MPK Runtime
#628 opened Feb 23, 2026 by xinhaoc • Draft
1 of 3 tasks
Add load_mpk for kernel reuse online-serving
#618 opened Feb 6, 2026 by AravTewari Loading…
Add fp8 gemm task on blackwell
#613 opened Jan 29, 2026 by yzhou442 • Draft
2 tasks
Add quantize fp8 task
#611 opened Jan 25, 2026 by yzhou442 Loading…
swizzling attn tasks to improve prefill performance
#608 opened Jan 17, 2026 by wenqiny • Draft
Multi turn support for vllm (based on vllm_integration)
#602 opened Dec 18, 2025 by NorthmanPKU Loading…
Removed -use_fast_math flag from compile command
#599 opened Dec 12, 2025 by AMKCode Loading…
Memory estimation for superoptimize
#597 opened Dec 12, 2025 by AMKCode Loading…
sm100 splitk demo added
#596 opened Dec 12, 2025 by JackFram Loading…
Program Partitioning
#592 opened Dec 6, 2025 by AMKCode Loading…
Blackwell MoE support
#571 opened Nov 12, 2025 by JackFram • Draft
[Not to land] support clang as compiler for mpk
#542 opened Oct 21, 2025 by wenqiny • Draft
Hopper: Add TMA load/store for Epilogue
#540 opened Oct 20, 2025 by xinhaoc • Draft
Software Pipelining Intra-SM
#510 opened Oct 1, 2025 by xinhaoc • Draft
Refactor norm_linear into struct functor
#425 opened Jul 24, 2025 by NorthmanPKU Loading…
6
Qwen3 demo batch size support for non-mirage baseline
#403 opened Jul 15, 2025 by dcw02 Loading…
5
start implementing llama 3.2 persistent kernel demo
#395 opened Jul 10, 2025 by dcw02 • Draft
8
Qwen3 Modal Demos
#394 opened Jul 10, 2025 by dcw02 • Draft
1
feat: Add sine and cosine operations to Mirage framework CUDA Transpiler Issues and features related to the CUDA transpiler of Mirage
#380 opened Jul 3, 2025 by EpicUsaMan Loading…
3
kernel level rowwise fp8 quantization; demo_hopper redo
#256 opened Apr 29, 2025 by Theorem411 Loading…
demo_hopper
#234 opened Mar 31, 2025 by Theorem411 Loading…
1
Adding operators with more than one output tensors: chunk op
#219 opened Mar 24, 2025 by AMKCode Loading…
2
Grace Hopper: Add threadblock cluster level reduction
#182 opened Feb 14, 2025 by xinhaoc • Draft
ProTip! Updated in the last three days: updated:>2026-03-03.