Code datasets for pretraining
Orion
Orion-zhen
AI & ML interests
Eco-friendly training using Tesla P4. Prefers (FSDP+)QLoRA.
Recent Activity
updated a Space about 2 months ago
Orion-zhen/tokenize-it updated a Space 2 months ago
Orion-zhen/vertin-tts updated a dataset 2 months ago
Orion-zhen/vertin-voicesOrganizations
spaces 6
pinned
Running
5
Tokenize It
😇
Visualize Hugging Face tokenizers in your browser
Running
6
Vertin TTS
🎙
DotTTS voice-cloning demo powered by audio.cpp
Paused
1
GGUF API
🌍
Serve an OpenAI compatible API on HF Space!
Paused
Space Fetch
🏆
Check computer statistics of your huggingface space
Paused
8
Deep Research
🌍
Use LLM for deep research
Runtime error
Agents
3
Qwen2.5 Math 7B Instruct GGUF
📈
Runs Qwen2.5-Math-7B-Instruct-GGUF
models 81
Orion-zhen/DeepSeek-R1-Distill-Qwen-32B-abliterated-AWQ
33B • Updated • 140 • 2
Orion-zhen/Ewen3-4b-Instruct-2507
4B • Updated • 7 • 3
Orion-zhen/Qwen3-30B-A3B-Instruct-2507-IQK-GGUF
31B • Updated • 186 • 1
Orion-zhen/Qwen3-30B-A3B-Thinking-2507-1M-GGUF
31B • Updated • 246
Orion-zhen/Dhanishtha-2.0-preview-0825-Q6_K-GGUF
Text Generation • 15B • Updated • 22
Orion-zhen/Dhanishtha-2.0-preview-0825-Q4_K_M-GGUF
Text Generation • 15B • Updated • 25
Orion-zhen/Qwen3-30B-A3B-Instruct-2507-1M-GGUF
31B • Updated • 283 • 1
Orion-zhen/Qwen3-30B-A3B-Thinking-2507-Q4_K_M-GGUF
31B • Updated • 81
Orion-zhen/Qwen3-30B-A3B-Instruct-2507-Q4_K_M-GGUF
31B • Updated • 212 • 2
Orion-zhen/Mellum-4B-Base-AWQ
4B • Updated • 16
datasets 28
Orion-zhen/vertin-voices
Viewer • Updated • 5 • 66 • 1
Orion-zhen/harmful-data
Viewer • Updated • 534 • 15
Orion-zhen/fim-code
Viewer • Updated • 616k • 64 • 1
Orion-zhen/github-python-code-fim
Viewer • Updated • 100k • 20
Orion-zhen/gsm8k-r1-qwen-32b
Viewer • Updated • 6.15k • 372 • 2
Orion-zhen/haiku-zh
Viewer • Updated • 3.93k • 19 • 2
Orion-zhen/word-count
Viewer • Updated • 48.7M • 16
Orion-zhen/meissa-lima
Viewer • Updated • 2.11k • 20
Orion-zhen/meissa-unalignments
Viewer • Updated • 66.3k • 37 • 8
Orion-zhen/Synthetic-Dark-RP-llama-factory
Viewer • Updated • 5 • 6 • 3