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

ProCreations/grug-27b-mtp-gguf overview

grug 27b mtp gguf grug rock WITH prediction head inside. Q8 0 / Q5 K M / Q4 K M, mtp tensors included for llama.cpp speculative decoding needs build with qwen3…

ggufgrugmtpspeculative-decodingenbase_model:ProCreations/grug-27b-mtpbase_model:quantized:ProCreations/grug-27b-mtplicense:apache-2.0endpoints_compatibleregion:usconversational

Runs locally from ~15.66 GB disk (16 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
0
Likes
3
Pipeline

Repository Files & Downloads

3 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
grug-27b-mtp-Q4_K_M.ggufGGUFQ4_K_M15.66 GBDownload
grug-27b-mtp-Q5_K_M.ggufGGUFQ5_K_M18.19 GBDownload
grug-27b-mtp-Q8_0.ggufGGUFQ8_027.05 GBDownload

Model Details

Model IDProCreations/grug-27b-mtp-gguf
AuthorProCreations
Pipeline
Licenseapache-2.0
Base modelProCreations/grug-27b-mtp
Last modified2026-07-23T19:40:52.000Z

Model README

---

license: apache-2.0

base_model: ProCreations/grug-27b-mtp

tags: [grug, mtp, gguf, speculative-decoding]

language: [en]

---

grug-27b-mtp-gguf

grug rock WITH prediction head inside. Q8_0 / Q5_K_M / Q4_K_M, mtp tensors

included for llama.cpp speculative decoding (needs build with qwen3_5 MTP

support). draft head trained on grug data: t+2 agreement 89.3%.

main card: grug-27b-mtp.

no-MTP rocks: grug-27b-gguf.

grug made by ProCreations.

Run ProCreations/grug-27b-mtp-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