alibaba--mnn

Files

T

History

Jbyang daae35c043 [Tools:Feature] llm_bench: add '-fa' meaning whether to use flash attention(default 1, using flash attention) (#4658 )

Discussed-in: Merge-Request 28749507 , URL: https://code.alibaba-inc.com/AliNN/AliNNPrivate/codereview/28749507
GitOrigin-RevId: 6f6b024d4e4c3f23a483aed5e7d0443102295a6f

Signed-off-by: jingbang.yjb <jingbang.yjb@alibaba-inc.com>

2026-07-22 16:28:52 +08:00

_static

MNN:Sync: Sync Internal 3.1.1

2025-03-12 11:35:16 +08:00

compile

[Metal:Speed] Fused Q4/Q8 dequant+GEMM kernel + M_TILE=64 variant for LLM prefill on M5 (#4651 )

2026-07-21 19:36:35 +08:00

contribute

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

cpp

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

inference

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

intro

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

perf

* [Metal:Speed] LLM prefill/decode kernel + shape/RoPE infrastructure + Metal Op Profile. (#4650 )

2026-07-21 19:32:55 +08:00

pymnn

Merge branch 'master' into update-docs

2025-07-28 09:42:11 +08:00

start

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

tools

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

train

[Doc:Refact] Restructure documentation for better navigation and newcomer experience

2026-03-25 12:50:30 +08:00

transformers

[Tools:Feature] llm_bench: add '-fa' meaning whether to use flash attention(default 1, using flash attention) (#4658 )

2026-07-22 16:28:52 +08:00

conf.py

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

faq.md

[Infra:Test] Add GEMV bandwidth benchmark for CPU/Metal w8/w4/w3/w2 (#4518 )

2026-06-05 18:15:13 +08:00

index.rst

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

make.bat

Sync Internal 2.0.5

2022-08-23 21:21:29 +08:00

Makefile

Project import generated by Copybara.

2025-12-22 09:53:32 +08:00

README

Sync Internal 2.0.5

2022-08-23 21:21:29 +08:00

requirements.txt

Docs:Bugfix: Update req sphinx to 5.0.0

2024-06-03 15:10:25 +08:00

testing.md

[Infra:Chore] Consolidate CI test scripts and docs (#4559 )

2026-06-23 13:06:40 +08:00

README

pip install -r requirements.txt
make html
open _build/html/index.html