LLM EXPLORER 61,491 MODELS INDEXED

Mahakal AI Fusion V3.0 by UX4567

By UX4567 · 225 downloads
HF Repo ↗ Copy URL

Mahakal AI Fusion V3.0 is an open-source language model by UX4567. Features: 3b LLM, VRAM: 6.2GB, Context: 32K, Instruction-Based, Merged, LLM Explorer Score: 0.31.

  Merged Model   Arxiv:2306.01708 Base model:qwen/qwen2.5-3b-ins... Base model:ux4567/kartik-kundl... Base model:ux4567/king-ai-chat...   Conversational   Endpoints compatible   Instruct   Qwen2   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/UX4567/Mahakal-AI-Fusion-v3.0 

Mahakal AI Fusion V3.0 Parameters and Internals

LLM NameMahakal AI Fusion V3.0
Repository 🤗https://huggingface.co/UX4567/Mahakal-AI-Fusion-v3.0 
Base Model(s)  UX4567/King-AI-Chat-v3.0   UX4567/Kartik-Kundli-AI-3B-v2.0   Qwen/Qwen2.5-3B-Instruct   UX4567/King-AI-Chat-v3.0   UX4567/Kartik-Kundli-AI-3B-v2.0   Qwen/Qwen2.5-3B-Instruct
Merged ModelYes
Model Size3b
Required VRAM6.2 GB
Updated2026-09-17
MaintainerUX4567
Model Typeqwen2
Instruction-BasedYes
Model Files  5.0 GB: 1-of-2   1.2 GB: 2-of-2
Model ArchitectureQwen2ForCausalLM
Context Length32768
Model Max Length32768
Transformers Version5.12.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Errorsreplace

Best Alternatives to Mahakal AI Fusion V3.0

Best Alternatives
Context / RAM
Downloads
Likes
VibeThinker 3B128K / 6.2 GB52745806
VibeThinker 3B128K / 6.2 GB3610
Saba2 3B128K / 6.2 GB60
Tessa T1 3B117K / 6.2 GB105
UIGEN T1.5 3B117K / 6.2 GB71
Qwen2.5 3B Instruct32K / 6.2 GB5889234545
King AI Chat V3.032K / 6.2 GB4871
...Instruct Sheldon SFT V2 Merged32K / 6.2 GB2160
...er 3B Instruct Spider Sft Grpo32K / 6.2 GB10251
SmallThinker 3B Preview32K / 6.8 GB30529412
Note: green Score (e.g. "73.2") means that the model is better than UX4567/Mahakal-AI-Fusion-v3.0.