Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
steveharing1
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
10 ms
·
61.
▲
by
steveharing1
5mo ago
The documentation is really helpful enough to get started
62.
▲
by
steveharing1
5mo ago
Neat local first setup to control Codex on your Mac from an iPhone. Uses a paired encrypted session and keeps everything self hosted.
63.
▲
Remodex: Control Codex from Your iPhone
(github.com)
1 points
by
steveharing1
5mo ago
|
1 comments
64.
▲
by
steveharing1
5mo ago
Can u suggest as per your experience what model you use and is it giving you intermediate to best coding results in your setup?
65.
▲
AI Coding Models You Can Run Locally on Consumer Hardware
(firethering.com)
2 points
by
steveharing1
5mo ago
|
1 comments
66.
▲
by
steveharing1
5mo ago
nice work!
67.
▲
by
steveharing1
5mo ago
I really appreciate open source community for moving this fast
68.
▲
Open Design: Use Your Coding Agent as a Design Engine
(github.com)
232 points
by
steveharing1
5mo ago
|
92 comments
69.
▲
by
steveharing1
5mo ago
yes. I've used it a lot. its very simple and good
70.
▲
by
steveharing1
5mo ago
You can try Open WebUI. Its genuinely useful when it comes to running open models locally with a clean interface
71.
▲
by
steveharing1
5mo ago
even calling it roll of the dice is an assumption. Can you point anything you find as mistake?
72.
▲
by
steveharing1
5mo ago
Thanks for letting me know
73.
▲
by
steveharing1
5mo ago
Because Qwen 3.6 pushes way above its weight. Granite 8B is impressive, but Qwen still wins on raw capability, especially for coding.
74.
▲
by
steveharing1
5mo ago
Yea, No doubt Qwen 3.6 open weights are far more strong
75.
▲
Granite 4.1: IBM's 8B Model Matching 32B MoE
(firethering.com)
328 points
by
steveharing1
5mo ago
|
209 comments
76.
▲
Laguna XS.2: A Coding Model Built for Closed Environments, Now Open Weights
(firethering.com)
1 points
by
steveharing1
5mo ago
|
0 comments
77.
▲
by
steveharing1
5mo ago
I think github is at a point that its too hard to ignore just like google is even though we might not like what they are doing now but we were the one made them this big.
78.
▲
Text-to-CAD: Generate 3D models with coding agents (open source)
(github.com)
2 points
by
steveharing1
5mo ago
|
0 comments
79.
▲
by
steveharing1
6mo ago
Lets see when the community come up with more quantized versions. Waiting for Unsloth's version
80.
▲
by
steveharing1
6mo ago
It was released around a week ago but Xiaomi Open Sourced it Now & its MIT Licensed and available om HuggingFace
81.
▲
Xiaomi releases MiMo-v2.5 Family weights with strong coding and agent benchmarks
(firethering.com)
21 points
by
steveharing1
6mo ago
|
7 comments
82.
▲
by
steveharing1
6mo ago
I won't say He should be working on it no matter what but I believe its a very good project and I think as always community forks will be the only option when it won't work in future
83.
▲
Open CoDesign: Open-source, local-first alternative to Claude Design and v0
(firethering.com)
11 points
by
steveharing1
6mo ago
|
1 comments
84.
▲
UniGenDet: Training image generators and detectors together
(ivg-yanranzhang.github.io)
1 points
by
steveharing1
6mo ago
|
0 comments
85.
▲
by
steveharing1
6mo ago
This one seems really impressive acc to bench scores but for me GLM 5.1 is still on top of every other open model so far
86.
▲
DeepSeek-V4: Making 1M token context efficient
(firethering.com)
3 points
by
steveharing1
6mo ago
|
0 comments
87.
▲
Prax: An agent runtime that learns from past mistakes and fixes code in a loop
(github.com)
4 points
by
steveharing1
6mo ago
|
0 comments
88.
▲
by
steveharing1
6mo ago
Yeah RYS is basically a no training way to test this idea
89.
▲
OpenMythos: A looped transformer take on how Claude Mythos might work
(firethering.com)
3 points
by
steveharing1
6mo ago
|
2 comments
90.
▲
Anthropic's relationship with the Trump administration seems to be thawing
(techcrunch.com)
1 points
by
steveharing1
6mo ago
|
0 comments
More ›