LLM EXPLORER 61,249 MODELS INDEXED

Nomi 1 Creative by LazyLoopStudio

By LazyLoopStudio · 104 downloads
HF Repo ↗ Copy URL

Nomi 1 Creative is an open-source language model by LazyLoopStudio. Features: 3b LLM, VRAM: 6.2GB, Context: 32K, License: apache-2.0, Quantized, Fine-Tuned, Instruction-Based, LLM Explorer Score: 0.22.

  4bit   Adapter-transformers Base model:adapter:unsloth/qwe... Base model:unsloth/qwen2.5-3b-...   Conversational   Creative Dataset:databricks/databricks-...   En   Endpoints compatible   Finetuned   Gguf   Instruct   Llama.cpp   Lmstudio   Nomi   Nomi-1   Nomi-creative   Ollama   Q4   Quantized   Qwen   Qwen2   Qwen2.5   Region:us   Safetensors   Sharded   Storytelling   Tensorflow   Unsloth
Model Card on HF 🤗: https://huggingface.co/LazyLoopStudio/Nomi-1-Creative 

Nomi 1 Creative Parameters and Internals

LLM NameNomi 1 Creative
Repository 🤗https://huggingface.co/LazyLoopStudio/Nomi-1-Creative 
Base Model(s)  unsloth/Qwen2.5-3B-Instruct-bnb-4bit   unsloth/Qwen2.5-3B-Instruct-bnb-4bit
Model Size3b
Required VRAM6.2 GB
Updated2026-08-04
MaintainerLazyLoopStudio
Model Typeqwen2
Instruction-BasedYes
Model Files  1.9 GB   2.5 GB   3.3 GB   4.0 GB: 1-of-2   2.2 GB: 2-of-2
Supported Languagesen
GGUF QuantizationYes
Quantization Typeq4|4bit|gguf|q4_k
Model ArchitectureQwen2ForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Tokenizer ClassQwen2Tokenizer
Padding Token<|vision_pad|>
Vocabulary Size151936
Torch Data Typefloat16
Errorsreplace

Best Alternatives to Nomi 1 Creative

Best Alternatives
Context / RAM
Downloads
Likes
VibeThinker 3B Heretic128K / 6.2 GB10164
TINY THINKER INSTRUCT32K / 6.2 GB13824
Spyfood Brain V232K / 6.2 GB1530
Zenith 3B32K / 6.3 GB1903
Qwen2.5 3B Instruct Heretic32K / 6.2 GB5651
Coder GRPO 3B32K / 6.2 GB2987
OpenCogito32K / 6.2 GB7790
OpenCognito R132K / 6.1 GB6550
Tiny Agent A 3B32K / 6.2 GB52624
Esbieta Ec Qwen 2.5 3B32K / 6.2 GB163
Note: green Score (e.g. "73.2") means that the model is better than LazyLoopStudio/Nomi-1-Creative.