Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Login
Top
New
Best
Ask
Show
Jobs
1.
▲
Show HN: Piqc – GPU waste scanner for LLM inference clusters
(github.com/paralleliq)
3 points
paralleliq
2 days ago
discuss
2.
▲
Telegram: 500m Active Users and a Noble Growth Plan
(gist.github.com)
1 point
parallelist
5 years ago
discuss
3.
▲
JavaScript tool to work with subtitles without pain
(github.com/termoxin)
2 points
termoxin
6 years ago
1 comment
4.
▲
Speed up Django commands through Queryset parallelization
(github.com/pelletier)
2 points
coulix
14 years ago
discuss
5.
▲
Build a LLM from Scratch – Technical Tutorial
(github.com/hundredblocks)
3 points
tf_12
3 years ago
discuss
6.
▲
Local working implementations of distributed ML model training
(github.com/hundredblocks)
3 points
e_ameisen
3 years ago
discuss
7.
▲
Parallelizing HNSW (Hierarchical Navigable Small World) graphs
(github.com/GavinMendelGleason)
49 points
LukeEF
2 years ago
6 comments
8.
▲
Speedup Megatron-LM from 10% to 30% with Zero Bubble Pipeline Parallelism
(github.com/sail-sg)
4 points
yzh
2 years ago
2 comments
9.
▲
Show HN: Asyncpal: Preemptive concurrency and parallelism for sporadic workloads
(github.com/pyrustic)
28 points
alexrustic
2 years ago
8 comments
10.
▲
Spelled out implementation of LLM parallelisms in pure C
(github.com/EugenHotaj)
3 points
eugenhotaj
2 years ago
discuss
11.
▲
Spice: Fine-grained parallelism with sub-nanosecond overhead in Zig
(github.com/judofyr)
394 points
dsp_person
2 years ago
53 comments
12.
▲
Show HN: Chili. Rust port of Spice, a low-overhead parallelization library
(github.com/dragostis)
233 points
dragostis
2 years ago
42 comments
13.
▲
DualPipe: Bidirectional pipeline parallelism algorithm
(github.com/deepseek-ai)
180 points
mfiguiere
a year ago
20 comments
14.
▲
Show HN: Forkrun – NUMA-aware shell parallelizer (50×–400× faster than parallel)
(github.com/jkool702)
151 points
jkool702
2 months ago
41 comments
15.
▲
YaFSDP: a sharded data parallelism framework, faster for pre-training LLMs
(github.com/yandex)
135 points
wiradikusuma
2 years ago
16 comments
16.
▲
Show HN: Coros – A Modern C++ Library for Task Parallelism
(github.com/mtmucha)
129 points
singledigits
2 years ago
58 comments
17.
▲
FBG – Lightweight C Linux framebuffer graphics with parallelism
(github.com/grz0zrg)
128 points
onirom
8 years ago
49 comments
18.
▲
DeepSeek Open Source Optimized Parallelism Strategies, 3 repos
(github.com/deepseek-ai)
103 points
helloericsf
a year ago
8 comments
19.
▲
Show HN: Parallelise pipelines of Python async iterables
(github.com/michalc)
71 points
michalc
6 years ago
41 comments
20.
▲
A Git helper tool that breaks large merges into parallelizable tasks
(github.com/mwallner)
68 points
schusterfredl
2 months ago
17 comments
21.
▲
EMS: Shared Memory Parallelism for Node.js
(github.com/syntheticsemantics)
59 points
lkgnmlkrewmgre
10 years ago
16 comments
22.
▲
Rayon: A data parallelism library for Rust
(github.com/rayon-rs)
48 points
nethunters
5 years ago
discuss
23.
▲
Juicesync: Parallelized Rsync for Cloud Storage
(github.com/juicedata)
28 points
daviesliu
6 years ago
7 comments
24.
▲
Show HN: Axe - A Systems Programming Language with Builtin Parallelism and No GC
(axelang.org)
16 points
death_eternal
6 months ago
5 comments
25.
▲
Show HN: Paralyze – a go package that simplifies parallelization
(github.com/i)
16 points
ilozinski
10 years ago
4 comments
26.
▲
GoPar: Automatic Loop Parallelization of Go Programs
(github.com/wetherbeei)
14 points
wetherbeei
13 years ago
3 comments
27.
▲
Forkrun – A pure-bash function for parallelizing loops
(github.com/jkool702)
9 points
shakna
2 years ago
discuss
28.
▲
Show HN: Agent Actors – Plan-Do-Check-Adjust with Parallelized LLM Agent Trees
(github.com/shaman-ai)
8 points
knrz
3 years ago
3 comments
29.
▲
Show HN: ChainFactory – Run Structured LLM Inference with Easy Parallelism
(github.com/pankajgarkoti)
8 points
garkotipankaj
2 years ago
discuss
30.
▲
Alpa: Auto-parallelizing large model training and inference (by UC Berkeley)
(github.com/alpa-projects)
7 points
zhisbug
4 years ago
1 comment
More