GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE →
Model Intelligence Sheet

creekhop/Qwen3.8-27b-EXP-EVE-GGUF overview

Qwen3.8 27b EXP EVE GGUF GGUF quantizations of: win10/Qwen3.8 27b EXP EVE https://huggingface.co/win10/Qwen3.8 27b EXP EVE Source Model Card This is a successf…

llama.cppggufqwenquantizedwin10Qwen3.8-27b-EXP-EVEbase_model:win10/Qwen3.8-27b-EXP-EVEbase_model:quantized:win10/Qwen3.8-27b-EXP-EVElicense:apache-2.0endpoints_compatibleregion:usconversational

Runs locally from ~15.02 GB disk (16 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
0
Likes
1
Pipeline
Author

Repository Files & Downloads

6 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
Qwen3.8-27b-EXP-EVE-BF16.ggufGGUFBF1650.90 GBDownload
Qwen3.8-27b-EXP-EVE-IQ4_NL.ggufGGUFIQ4_NL15.02 GBDownload
Qwen3.8-27b-EXP-EVE-Q4_K_M.ggufGGUFQ4_K_M15.66 GBDownload
Qwen3.8-27b-EXP-EVE-Q5_K_M.ggufGGUFQ5_K_M18.19 GBDownload
Qwen3.8-27b-EXP-EVE-Q6_K.ggufGGUFQ6_K20.89 GBDownload
Qwen3.8-27b-EXP-EVE-Q8_0.ggufGGUFQ8_027.05 GBDownload

Model Details

Model IDcreekhop/Qwen3.8-27b-EXP-EVE-GGUF
Authorcreekhop
Pipeline
Licenseapache-2.0
Base modelwin10/Qwen3.8-27b-EXP-EVE
Last modified2026-08-25T14:26:45.000Z

Model README

---

license: apache-2.0

base_model:

  • win10/Qwen3.8-27b-EXP-EVE

base_model_relation: quantized

library_name: llama.cpp

tags:

  • gguf
  • qwen
  • llama.cpp
  • quantized
  • win10
  • Qwen3.8-27b-EXP-EVE

---

Qwen3.8-27b-EXP-EVE-GGUF

GGUF quantizations of:

win10/Qwen3.8-27b-EXP-EVE

Source Model Card

This is a successfully merged cross-architecture model, built around the popular Qwen3.8-27B as its primary backbone.

Using a Tensor Gene Evolution merging approach, I incorporated capabilities and behavioral characteristics from meta-models/Muse-Glimmer-30B and google/gemma-4-31B-it into the Qwen backbone.

Compared with the original base model, this merged model appears to preserve the donors' reasoning characteristics more effectively, while demonstrating broader reasoning coverage, greater depth of thought, and more diverse problem-solving behavior.

In my observations, its reasoning ability can be surprisingly strong and, in some cases, may even appear more capable than DeepSeek models. Further systematic evaluation and benchmarking are still needed to quantify these differences.

💖 Open Sponsorship

Support this research and future model development:

Run creekhop/Qwen3.8-27b-EXP-EVE-GGUF with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models

Source: Hugging Face · Compare models