slotstream
AI
local llm inference
SHOW HN2026
240points
118comments
#210of 4,366 in 2026
Top 5%of 2026 by points
How HN took it
MixedNo clear lean
3 positive, 18 neutral or asking, 2 mixed, 17 critical, out of 40 comments about the project. Thread score -0.34.
Performance: 0 for, 7 againstDocs: 0 for, 6 againstIdea: 1 for, 4 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Interest in local models running on constrained hardware
From 2 comments
I'm hoping to see progress in this space. Folks talking about how 32G is not enough for local use, but then there's been work like this to empower it. My hope is that the new 32G…
whartung on HNI love these efforts to get proper models running on lower cost hardware and I think this is where the next real breakthrough will come from. The more efficient this sort of…
jacquesm on HN
Objections
Redundant with existing similar projects
From 2 comments
There are already a handful of repos doing essentially exactly this: `mlx-moe-offload`, `streamlx`, `mlx-moe`, `mlx-flash`, and `deepseek-v4-flash-mlx` - i.e. keep the resident…
AmazingTurtle on HNso many inference project, omlx already supports all of this and has a 1000 people trying to optimize it constantly
c0rruptbytes on HN
README reads like AI-generated text
From 2 comments
> Hugging Face is the bottleneck, not your link. README could clearly make use of a cleanup, seems to be more like a session log dump now than a good introduction to the project…
embedding-shape on HNI hate this AI style writing because since it doesn’t really understand flow, it’s being inserted in irrelevant places and it is extremely irritating to read.
Eufrat on HN
Disk reads may wear the SSD
From 2 comments
Is this going to destroy my SSD?
jonplackett on HNReading causes insignificant wear ("read disturb") that likely isn't a problem, but I don't think it's possible to issue pure reads to modern SSDs. The NVMe spec mandates…
mrob on HN
Requests
Lower memory use per process
From 1 comment
I just used it, went through the whole installation (took like 1 hour approx). Long but straightforward. If I put my computer to sleep will it continue? I started the server…
baristaGeek on HN
Benchmarks against other inference tools
From 2 comments
Would love to see metrics of model performance, comparison with oMLX/OLLAMA/others. Any tooling for local optimization on hardware.
securecloudgrou on HNI'm sorry this makes it seem like I didn't do my research. I did a TON. To fix it I'll add a benchmark/comparison table. Also, I wouldn't call it market research since this is…
carloslfu on HN
Native desktop app
From 1 comment
Also I don't know if people like this... but could this be wrapped around a nice desktop app?
baristaGeek on HN
Questions
Energy consumption
From 1 comment
how much energy does it consume?
ErenayDev on HN
Real use case
From 1 comment
I just used it, went through the whole installation (took like 1 hour approx). Long but straightforward. If I put my computer to sleep will it continue? I started the server…
baristaGeek on HN
Alternatives named
oMLX
From 1 comment
so many inference project, omlx already supports all of this and has a 1000 people trying to optimize it constantly
c0rruptbytes on HN
mlx-moe-offload, streamlx, mlx-flash
From 1 comment
There are already a handful of repos doing essentially exactly this: `mlx-moe-offload`, `streamlx`, `mlx-moe`, `mlx-flash`, and `deepseek-v4-flash-mlx` - i.e. keep the resident…
AmazingTurtle on HN
Ollama
From 1 comment
I just used it, went through the whole installation (took like 1 hour approx). Long but straightforward. If I put my computer to sleep will it continue? I started the server…
baristaGeek on HN