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

poisonxa/Qwable-27B-GGUF overview

Qwable 27B GGUF benchmark files The two files used for the PXA cross engine speed comparisons on Tesla P100 and V100 cards, published so every number in the PX…

ggufqwenpxqbenchmarklicense:apache-2.0endpoints_compatibleregion:usimatrixconversational

Runs locally from ~14.60 GB disk (16 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
569
Likes
0
Pipeline
Author

Repository Files & Downloads

2 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
Qwable-27B-MXFP4-lite.ggufGGUFGGUF14.60 GBDownload
Qwable-27B-PXQ4core.ggufGGUFGGUF14.60 GBDownload

Model Details

Model IDpoisonxa/Qwable-27B-GGUF
Authorpoisonxa
Pipeline
Licenseapache-2.0
Base model
Last modified2026-09-09T02:07:28.000Z

Model README

---

license: apache-2.0

tags: [qwen, gguf, pxq, benchmark]

---

Qwable-27B GGUF (benchmark files)

The two files used for the PXA cross-engine speed comparisons on Tesla P100 and V100 cards, published so every number in the

PXA release notes and bench/fair-battle.md can be reproduced from a download.

Both are conversions of the same 27B dense hybrid (Qwen3.8-27B lineage, abliterated), so the engines are compared on identical weights.

| File | Format | Size | sha256 (first 16) | Runs on |

|---|---|---|---|---|

| Qwable-27B-PXQ4core.gguf | PXQ4 (pxq_llama codec) | 15.7 GB | 6fb4a437684e8bbe | pxq_llama only |

| Qwable-27B-MXFP4-lite.gguf | MXFP4 | 15.0 GB | 1d04b75817358cf7 | mainline llama.cpp, ik_llama.cpp, pxq_llama |

Protocol: bench/fair/protocol.md in the engine repo (temperature 0, median of 7, sha-checked files, unique prompts).

Full sha256 sums are in checksums.sha256 here and in bench/fair/weights/MANIFEST.sha256 in the repo.

Community: Discord — PXA Network

Run poisonxa/Qwable-27B-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