license: apache-2.0
base_model: "orcarouter/Qwen3.8-27B-Uncensored-MLX"
tags:
- preservation-mirror
- mango
Preservation mirror: orcarouter/Qwen3.8-27B-Uncensored-MLX
Mirrored by isilicon from orcarouter/Qwen3.8-27B-Uncensored-MLX at commit a92a9d3492320048ca96924262d158f75ab5ce8b. Original training, edits and quantization belong to the upstream creators. This mirror does not claim new training or authorship.
Files retain their original selected paths and bytes. This repository contains the selected quantization/dependency, not every quantization from the source. See SOURCE_MANIFEST.json for every path, size and checksum, and UPSTREAM_README.md for original model information, attribution and caveats when supplied upstream. Original available license/notice files are retained. The source model card declares the license shown above; base-model conditions still apply.
Keep original subfolders when downloading (for example 4-bit/ for MLX). Tokenizers, projectors, draft models and offloaded tensor dependencies may live in separate mirrors. A weight-only mirror is not a claim of standalone runtime readiness.
Required companion mirrors
Native engines and their launch settings must also be restored separately.
Local Mango evaluation
Weighted score 80.64/100, 202 suite questions plus 12 agent cases. Mechanical agent passes: 12/12; recorded workload 23.98 minutes; infrastructure errors 0.
These measurements apply to this selected quantization with our recorded runtime and dependencies. They are local automatic/mechanical scores on a convenience sample, not an official universal leaderboard or a complete semantic audit. EVALUATION.json contains the exact label, domain/task scores, weighting, workload and limitations.
Preservation status
MIRROR_STATUS.json, when present, records destination-file verification against the pinned source metadata. This is not a fresh independent-device full restore/load test. An HF mirror remains dependent on Hugging Face availability and policies.