kenonix/Darwin-9B-KOREA-Q8_0-GGUF
Text Generation • 9B • Updated • 14 • 8
kenonix
kenonix
AI & ML interests
None yet
Recent Activity
liked a model 1 day ago
Comfy-Org/MOSS-Image liked a model 2 days ago
RecursiveSI/flint reacted to jiteshkprasad's post with 🔥 2 days ago
Last week we released Gargi, a 384M English language model trained from scratch on about 17B tokens, roughly 1,000× less training data than Qwen2.5-0.5B (~18T tokens, as published by its authors).
What we measured (same harness for every model: lm-evaluation-harness, 0-shot, greedy, each model's start token):
- Base model vs Qwen2.5-0.5B: ARC-Easy 62.8 vs 64.7, ARC-Challenge 32.3 vs 32.4, WinoGrande 54.6 vs 56.1.
- Chat model: thinking on or off; GSM8K 36.0 with thinking on.
- Long inputs: needle-in-a-haystack recall 100% at 8K, 93% at 16K and 90% at 20K tokens, so use it up to about 16K.
Where it falls short: knowledge and everyday common sense still scale with data. MMLU is 25.5 vs 47.3, and it trails on HellaSwag and PIQA. It is not a maths specialist, and it is not safety-tuned: put a moderation layer in front of any public use.
It runs on a laptop: GGUF (Q8_0 at 410 MB, Q4_K_M, f16) and MLX builds.
`ollama run hf.co/udgamlabs/Gargi-GGUF:Q8_0`
Apache-2.0 ·
weights: https://huggingface.co/udgamlabs/Gargi
try it: https://udgamlabs.com/models/gargi
code: https://github.com/udgamlabs/gargi
Named after Gargi Vachaknavi, the Vedic philosopher remembered for asking the hardest questions.Organizations
None yet
kenonix/Darwin-9B-KOREA-Q4_K_M-GGUF
Text Generation • 9B • Updated • 30 • 8
kenonix/AWAXIS-Think-28B-Q4_K_M-GGUF
Text Generation • 27B • Updated • 20 • 1
kenonix/AWAXIS-Think-28B-Q8_0-GGUF
Text Generation • 27B • Updated • 38 • 1
kenonix/Llama-3.3-8B-Thinking-Gemini-Flash-11000x-128k-Q8_0-GGUF
Text Generation • 8B • Updated • 69 • 1
kenonix/Llama-3.3-8B-Thinking-Gemini-Flash-11000x-128k-Q4_K_M-GGUF
Text Generation • 8B • Updated • 96 • 1
kenonix/gemma-3N-More-math-ko2-IQ4_XS-GGUF
7B • Updated • 9
kenonix/gemma-3N-More-math-ko2-Q5_K_M-GGUF
7B • Updated • 31
kenonix/gemma-3N-More-math-ko2-IQ4_NL-GGUF
7B • Updated • 10
kenonix/Kogemma3ne4b
Updated
kenonix/ko-gemma-3n-test
Updated
kenonix/Nix-ungemma-3n-E4B
Updated
kenonix/gemma-3N-More-math-ko2-Q4_0-GGUF
7B • Updated • 18
kenonix/gemma-3N-More-math-ko2-Q8_0-GGUF
7B • Updated • 6
kenonix/gemma-3N-More-math-ko2-Q4_K_M-GGUF
7B • Updated • 17
kenonix/gemma-3N-More-math-ko2-Q6_K-GGUF
7B • Updated • 18
kenonix/gemma-3N-More-math-ko2
Image-Text-to-Text • 8B • Updated • 5
kenonix/gemma-3N-More-math-ko-Q8_0-GGUF
7B • Updated • 17
kenonix/gemma-3N-More-math-ko-Q6_K-GGUF
7B • Updated • 20
kenonix/gemma-3N-More-math-ko-Q4_K_M-GGUF
7B • Updated • 10
kenonix/gemma-3N-More-math-ko
Image-Text-to-Text • 8B • Updated • 10
kenonix/gemma-3N-More-math
Image-Text-to-Text • 8B • Updated • 9
kenonix/next-4b-Q4_K_M-GGUF
Image-Text-to-Text • 4B • Updated • 10
kenonix/next-4b-Q5_0-GGUF
Image-Text-to-Text • 4B • Updated • 33
kenonix/next-4b-Q6_K-GGUF
Image-Text-to-Text • 4B • Updated • 29
kenonix/next-4b-Q8_0-GGUF
Image-Text-to-Text • 4B • Updated • 18
kenonix/qwen3-1.7B-ko-summary-finetuned-06-12-Q8_0-GGUF
Summarization • 2B • Updated • 8
kenonix/gemma-3n-E4B-IT-LoRA1
Updated
kenonix/gemma-3n-E4B-IT-LoRA
Updated
kenonix/gemma-3-ko-4B-1-rkllm
Updated