HK
Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Top
New
Best
Ask
Show
Jobs
Request
1.
▲
Context is essential for AI agents, but I think shared state is the next problem
github.com/Abloatai
1 comment
21 days ago
lukasanderss
2 points
2.
▲
Ablo: The Collaboration Layer for AI Agents
github.com/Abloatai
discuss
5 days ago
luckymonkybaby
1 points
3.
▲
Time Ablation Experiments on tau2-bench
github.com/sshh12
discuss
5 months ago
sshh12
3 points
4.
▲
Show HN: Toolkit for LLM Fine-Tuning, Ablating and Testing
github.com/georgian-io
21 comments
2 years ago
rsaha7
14 points
5.
▲
Show HN: Implementation and ablation of the Hierarchical Reasoning Model (HRM)
github.com/krychu
discuss
9 months ago
krychu
2 points
6.
▲
Show HN: Forge – Guardrails take an 8B model from 53% to 99% on agentic tasks
github.com/antoinezambelli
252 comments
a month ago
zambelli
687 points
7.
▲
Show HN: I Created ErisForge, a Python Library for Abliteration of LLMs
github.com/Tsadoq
52 comments
a year ago
tsadoq
140 points
8.
▲
Show HN: Desbordante 2.0 – A high-performance data profiler
github.com/Desbordante
14 comments
2 years ago
chernishev
67 points
9.
▲
Show HN: Chonky – a neural text semantic chunking goes multilingual
huggingface.co
6 comments
8 months ago
hessdalenlight
43 points
10.
▲
Show HN: A tiny reasoning layer that steadies LLM outputs (MIT; +22.4% accuracy)
github.com/onestardao
discuss
a year ago
TXTOS
6 points
11.
▲
HRNet; the most powerful new vision CNN architecture to come out since AlexNet!
discuss
7 years ago
timscarfe
3 points
12.
▲
Show HN: I built a 2nd-order PyTorch optimizer for LLMs that runs on 16GB GPUs
4 comments
2 months ago
dnosoz
2 points
13.
▲
Show HN: SoMatic – Vision-based OS automation framework for AI agents
github.com/Smyan1909
discuss
a month ago
smyansondur
2 points
14.
▲
Show HN: FT-Lab – Lightweight TinyLlama Fine-Tuning (Full FT / LoRA / QLoRA)
github.com/REICHIYAN
discuss
7 months ago
Sai-HN
2 points
15.
▲
MiRAGE: Open-source framework for multimodal RAG evaluation
discuss
4 months ago
mmhetric
1 points
16.
▲
Show HN: WLM – A 70B model trained to decode "I'm fine" with 94.7% accuracy
github.com/gabewillen
discuss
5 months ago
gwillen85
1 points
17.
▲
Show HN: FT-Lab – Lightweight TinyLlama Fine-Tuning (Full FT / LoRA / QLoRA)
github.com/REICHIYAN
discuss
7 months ago
Sai-HN
1 points
18.
▲
New QA system released open source with reviews
github.com/bernhard2202
discuss
7 years ago
ablation125
2 points