LLM EXPLORER 66,524 MODELS INDEXED

LittleKedi Medium Wide Sparse Instruct by edededdy

By edededdy · 132 downloads
HF Repo ↗ Copy URL

LittleKedi Medium Wide Sparse Instruct is an open-source language model by edededdy. Features: 305.4m LLM, VRAM: 0.3GB, License: apache-2.0, Quantized, Instruction-Based, LLM Explorer Score: 0.3.

Base model:edededdy/littlekedi... Base model:finetune:edededdy/l...   Chat   Conversational Dataset:codeparrot/github-code... Dataset:huggingfacefw/fineweb-... Dataset:huggingfacegeclm/reddi...   Dataset:huggingfacetb/finemath Dataset:huggingfacetb/smollm-c...   Dataset:huggingfacetb/smoltalk   Dataset:wikimedia/wikipedia   Deepseek   En   Endpoints compatible   From-scratch   Gguf   Instruct   Instruction-tuned   Littlekedi   Mixture-of-experts   Mla   Moe   Multi-head-latent-attention   Q8   Quantized   Region:us   Safetensors   Tiny
Model Card on HF 🤗: https://huggingface.co/edededdy/LittleKedi-medium-wide-sparse-instruct 

LittleKedi Medium Wide Sparse Instruct Parameters and Internals

LLM NameLittleKedi Medium Wide Sparse Instruct
Repository 🤗https://huggingface.co/edededdy/LittleKedi-medium-wide-sparse-instruct 
Base Model(s)  ...leKedi Medium Wide Sparse Base   edededdy/LittleKedi-medium-wide-sparse-base
Model Size305.4m
Required VRAM0.3 GB
Updated2026-11-03
Maintaineredededdy
Instruction-BasedYes
Model Files  0.6 GB   0.3 GB   1.2 GB
Supported Languagesen
GGUF QuantizationYes
Quantization Typeq8|gguf
Licenseapache-2.0
Model Max Length2048
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|endoftext|>
Vocabulary Size16384