LLM EXPLORER 66,383 MODELS INDEXED

Kvark Sparkle 1.0 27B by Egzakta

By Egzakta · 30 downloads
HF Repo ↗ Copy URL

Kvark Sparkle 1.0 27B is an open-source language model by Egzakta. Features: 27b LLM, VRAM: 54.7GB, License: apache-2.0, LLM Explorer Score: 0.27.

Base model:finetune:qwen/qwen3...   Base model:qwen/qwen3.8-27b   Conversational Dataset:allenai/if multi const... Dataset:livecodebench/code gen...   Dataset:m-a-p/supergpqa   Dataset:zwhe99/deepmath-103k   Efficient-reasoning   En   Endpoints compatible   Image-text-to-text   Kvark   Model-index   Mtp   Qwen3.8   Qwen3 5   Reasoning   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Egzakta/Kvark-Sparkle-1.0-27B 

Kvark Sparkle 1.0 27B Parameters and Internals

LLM NameKvark Sparkle 1.0 27B
Repository 🤗https://huggingface.co/Egzakta/Kvark-Sparkle-1.0-27B 
Base Model(s)  Qwen/Qwen3.8-27B   Qwen/Qwen3.8-27B
Model Size27b
Required VRAM54.7 GB
Updated2026-11-01
MaintainerEgzakta
Model Typeqwen3_5
Model Files  2.5 GB: 1-of-12   4.8 GB: 2-of-12   5.0 GB: 3-of-12   4.9 GB: 4-of-12   5.0 GB: 5-of-12   4.9 GB: 6-of-12   4.9 GB: 7-of-12   5.0 GB: 8-of-12   5.0 GB: 9-of-12   4.9 GB: 10-of-12   5.0 GB: 11-of-12   2.8 GB: 12-of-12   0.8 GB
Supported Languagesen
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.19.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Kvark Sparkle 1.0 27B

Best Alternatives
Context / RAM
Downloads
Likes
...wopus3.8 27B Flash Mlx OQ4e 1M1024K / 16.9 GB2111
Agens Pilot1024K / 54.7 GB863
Qwen3.8 27B NPU2256K /  GB2524
JEV 27B VL0K / 55.6 GB1525286966
Clef0K / 54.7 GB220432237
Qwen3.8 27B NVFP4 DGX SPARK V20K / 19.8 GB11317113
Qwen3.6 27B0K / 55.4 GB67281382197
Qwen3.8 27B0K / 55.6 GB28753
Qwen3.8 27B0K / 55.6 GB272721
Qwen3.8 27B MLX0K / 16 GB141118
Note: green Score (e.g. "73.2") means that the model is better than Egzakta/Kvark-Sparkle-1.0-27B.