LLM EXPLORER 60,220 MODELS INDEXED

Velvet Eclipse 4x12B V0.2 by SuperbEmphasis

By SuperbEmphasis · 8 downloads
HF Repo ↗ Copy URL

Velvet Eclipse 4x12B V0.2 is an open-source language model by SuperbEmphasis. Features: 38.7b LLM, VRAM: 77.7GB, Context: 1000K, MoE, LLM Explorer Score: 0.18.

  Creative Dataset:superbemphasis/claude-... Dataset:superbemphasis/deepsee... Dataset:superbemphasis/deepsee...   Mergekit   Mixtral   Moe   Region:us   Roleplaying   Rp   Safetensors   Sharded   Story   Storywriting   Tensorflow   Writing
Model Card on HF 🤗: https://huggingface.co/SuperbEmphasis/Velvet-Eclipse-4x12B-v0.2 

Velvet Eclipse 4x12B V0.2 Parameters and Internals

LLM NameVelvet Eclipse 4x12B V0.2
Repository 🤗https://huggingface.co/SuperbEmphasis/Velvet-Eclipse-4x12B-v0.2 
Model Size38.7b
Required VRAM77.7 GB
Updated2026-06-22
MaintainerSuperbEmphasis
Model Typemixtral
Model Files  4.9 GB: 1-of-16   4.9 GB: 2-of-16   4.9 GB: 3-of-16   4.9 GB: 4-of-16   4.9 GB: 5-of-16   5.0 GB: 6-of-16   5.0 GB: 7-of-16   4.9 GB: 8-of-16   4.9 GB: 9-of-16   4.9 GB: 10-of-16   4.9 GB: 11-of-16   4.9 GB: 12-of-16   4.9 GB: 13-of-16   5.0 GB: 14-of-16   5.0 GB: 15-of-16   3.8 GB: 16-of-16
Model ArchitectureMixtralForCausalLM
Context Length1024000
Model Max Length1024000
Transformers Version4.52.4
Tokenizer ClassPreTrainedTokenizer
Padding Token<s>
Vocabulary Size131072
Torch Data Typebfloat16