Public HSL/HoLo evidence: byte-native inputs, language models, speech, diffusion, weights, demos, matched controls, and honest limits.
Jinhyun Woo
ggunio
AI & ML interests
None yet
Recent Activity
updated a collection 2 days ago
HSL / HoLo - Open Feasibility Portfolio updated a collection 2 days ago
HSL / HoLo - Open Feasibility Portfolio updated a collection 2 days ago
HSL / HoLo - Open Feasibility PortfolioOrganizations
None yet
spaces 7
pinned
Paused
Agents
1
B2NL v6.2.1 - Byte-to-Natural Language Tokenizer 🚀
🚀
Compress and reconstruct text into semantic embeddings
Running on Zero
Agents
HoLo-FuSe
🐱
0-param frozen HSL substrate conditioning a 128px DDPM
Sleeping
Agents
HoLo-ToLk
🗣
Tokenizer-free speech line: STT + TTS on HSL
Sleeping
Agents
HoLo_ZeRo
🌀
Byte transformer, 0 learned input params, I/O-symmetric
Sleeping
Agents
HoLo · Research POC
🌌
Change-rate-based byte-native multimodal AI (POC)
Sleeping
Agents
HSL Embedding Demo
🌊
Upload any file - HSL (27-D) embeds & rebuilds it exactly
models 9
ggunio/HoLo-FuSe
Unconditional Image Generation • Updated
ggunio/HoLo-ToLk-TTS
Text-to-Speech • Updated • 1
ggunio/HoLo-ToLk-STT
Automatic Speech Recognition • Updated
ggunio/HoLo_ZeRo
Updated
ggunio/HoLo-6.5.1
Updated
ggunio/B2NL-IntelligentTokenizer-v6.2.1
0.2B • Updated • 17
ggunio/B2NL-v6.1.1
Updated • 3
ggunio/B2NL-v6.1.2
Updated • 6
ggunio/intelligent-tokenizer-v6
Updated • 3
datasets 0
None public yet