LLM EXPLORER 59,420 MODELS INDEXED

Buttercup 4x7B 6bpw EXL2 by Kooten

By Kooten · 6 downloads
HF Repo ↗ Copy URL

Buttercup 4x7B 6bpw EXL2 is an open-source language model by Kooten. Features: 7b LLM, VRAM: 18.4GB, Context: 32K, License: apache-2.0, MoE, Quantized, LLM Explorer Score: 0.11.

  En   Endpoints compatible   Exl2   Merge   Mixtral   Moe   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Kooten/Buttercup-4x7B-6bpw-exl2 

Buttercup 4x7B 6bpw EXL2 Parameters and Internals

Model Type 
moe, merge
Additional Notes 
Contact Kooten on discord.
Supported Languages 
en (High)
Input Output 
Input Format:
Unclear
LLM NameButtercup 4x7B 6bpw EXL2
Repository 🤗https://huggingface.co/Kooten/Buttercup-4x7B-6bpw-exl2 
Model Size7b
Required VRAM18.4 GB
Updated2026-08-04
MaintainerKooten
Model Typemixtral
Model Files  8.6 GB: 1-of-3   8.6 GB: 2-of-3   1.2 GB: 3-of-3
Supported Languagesen
Quantization Typeexl2
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.37.0.dev0
Tokenizer ClassLlamaTokenizer
Padding Token<s>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to Buttercup 4x7B 6bpw EXL2

Best Alternatives
Context / RAM
Downloads
Likes
...izardLM 2 4x7B MoE EXL2 3 0bpw32K / 9.4 GB21
BurningBruce 003 EXL2 B8.032K / 24.3 GB71
...ixtral 2x7b DPO 8.0bpw H8 EXL232K / 13 GB44
...ixtral 2x7b DPO 6.0bpw H6 EXL232K / 9.8 GB61
...ixtral 2x7b DPO 4.0bpw H6 EXL232K / 6.7 GB21
...ixtral 2x7b DPO 5.0bpw H6 EXL232K / 8.3 GB11
...t V0.2 2x7B MoE 6.0bpw H6 EXL232K / 9.9 GB31
...iveFusion 4x7b MoE EXL2 3.5BPW8K / 10.8 GB81
...lue Orchid 2x7b 8.0bpw H8 EXL28K / 13 GB75
...lue Orchid 2x7b 5.0bpw H6 EXL28K / 8.2 GB84
Note: green Score (e.g. "73.2") means that the model is better than Kooten/Buttercup-4x7B-6bpw-exl2.