LLM EXPLORER 59,071 MODELS INDEXED

Grug 35B A3B by kai-os

By kai-os · 1252 downloads
HF Repo ↗ Copy URL

Grug 35B A3B is an open-source language model by kai-os. Features: 34.7b LLM, VRAM: 69GB, Context: 256K, License: apache-2.0, MoE, Instruction-Based, LLM Explorer Score: 0.3.

  A3b Base model:finetune:qwen/qwen3... Base model:qwen/qwen3.6-35b-a3...   Compact-reasoning   Conversational Dataset:cl-from-nothing/code r...   Dataset:hotdogs/uka-glm-5.2 Dataset:hsh-intelligence/verif... Dataset:kd13/codedebug-instruc... Dataset:madarabr/cortex-adapti... Dataset:samcheng0/lumia-reason... Dataset:scale-or-reason/genera...   Endpoints compatible   Finetuned   Instruct   Model-index   Moe   Qlora   Qwen   Qwen3.6   Qwen3 5 moe text   Reasoning   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/kai-os/Grug-35B-A3B 

Grug 35B A3B Parameters and Internals

LLM NameGrug 35B A3B
Repository 🤗https://huggingface.co/kai-os/Grug-35B-A3B 
Base Model(s)  Qwen/Qwen3.6-35B-A3B   Qwen/Qwen3.6-35B-A3B
Model Size34.7b
Required VRAM69 GB
Updated2026-07-29
Maintainerkai-os
Model Typeqwen3_5_moe_text
Instruction-BasedYes
Model Files  3.8 GB: 1-of-21   3.8 GB: 2-of-21   3.4 GB: 3-of-21   3.3 GB: 4-of-21   3.4 GB: 5-of-21   3.3 GB: 6-of-21   3.4 GB: 7-of-21   3.3 GB: 8-of-21   3.4 GB: 9-of-21   3.3 GB: 10-of-21   3.4 GB: 11-of-21   3.3 GB: 12-of-21   3.4 GB: 13-of-21   3.3 GB: 14-of-21   3.4 GB: 15-of-21   3.3 GB: 16-of-21   3.4 GB: 17-of-21   3.3 GB: 18-of-21   3.4 GB: 19-of-21   3.3 GB: 20-of-21   1.1 GB: 21-of-21
Model ArchitectureQwen3_5MoeForCausalLM
Licenseapache-2.0
Context Length262144
Model Max Length262144
Transformers Version5.12.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size248320
Errorsreplace

Best Alternatives to Grug 35B A3B

Best Alternatives
Context / RAM
Downloads
Likes
Aurora Code 1256K / 69 GB4435