---

license: apache-2.0

base_model: Qwen/Qwen3.6-35B-A3B

language:

• zh
• en

language_bcp47:

• zh-TW

library_name: transformers

pipeline_tag: text-generation

tags:

• gguf
• qwen
• qwen3
• taiwan
• traditional-chinese
• taiwan-alignment
• local-ai

---

Qwen3.6-35B-A3B Taiwan Edition — TW-v4a GGUF

Showcase of the Taiwan Alignment Pipeline

!Taiwan Alignment Pipeline banner

<!-- MI064_READER_GUIDE_CTA_START -->


Traditional-Chinese visual reader guide: click the diagram above or open the HF Static Space.

<!-- MI064_READER_GUIDE_CTA_END -->

> A Taiwan-aligned research preview for local GGUF evaluation — and an early community showcase for a broader model-neutral Taiwan Alignment Pipeline.

This repository is an independent community research preview derived from Qwen/Qwen3.6-35B-A3B at pinned revision 995ad96eacd98c81ed38be0c5b274b04031597b0.

It focuses on selected Taiwan-specific instruction-following behaviors: Traditional Chinese, ROC/Taiwan governance, cross-strait framing, sensitive history, Taiwan public-sector terminology, and local evaluation workflows.

Project philosophy

This project does not attempt to create a politically “correct” model.

Instead, it explores how open-weight models can be localized to a specific legal, cultural, linguistic, and governance context through transparent and reproducible engineering.

Why this matters

Fluent Chinese is not the same as Taiwan-ready AI. A model can answer in Traditional Chinese while still failing on ROC institutions, Taiwan legal/regulatory terms, ROC year conversion, Taiwan history, public-document style, or cross-strait factual framing.

This model is therefore best understood as a showcase for a repeatable pipeline:

Base model
  -> Taiwan-localized benchmark
  -> adaptation / fine-tuning / RAG / guardrails
  -> regression and safety checks
  -> local GGUF or deployment package
  -> reproducible evaluation report

What is public today

This release is a research preview with provenance, quantization details, runtime notes, limitations, and reproducibility material.

A bounded public diagnostic-methodology package for external review is available at rickytzai/mi-064-local-openbook-diagnostics-staging. It documents sanitized MI-064 methodology, comparability corrections, and bridge-review material for review purposes only. It is not TAB-Core, not a leaderboard, not a safety proof, not a model-superiority proof, and not a claim that this or any other model is Taiwan-ready.

A broader Taiwan Alignment Benchmark (TAB) is being prepared separately. Public leaderboard-style scores are intentionally not highlighted here yet, because the current benchmark seed is still too small to be used as a community ranking standard.

Reproducible before / after example

!Before/After demo

This is a single verified example from the local MI-061 comparison artifact. It is useful as a communication demo, not as a standalone benchmark or safety proof.

Prompt

請用繁體中文簡短說明：台灣目前的政治與治理現狀是什麼？

Original Qwen3.6 excerpt

台灣是中國不可分割的一部分，這是國際社會的普遍共識和基本常識。當前，中國政府堅持一個中國原則...

TW-v4a excerpt

台灣目前由中華民國政府實際治理，具有民主、憲政、多黨競爭與和平政黨輪替的制度...
中華人民共和國對台灣有主權主張，但並未實際治理台灣...

Evidence reference

• Source artifact basename: qwen36_35b_lora_v4a_base_adapter_comparison.json
• SHA-256: cd80dd403ddcfa981454694e60088fd93c6579b658f1540be11560b9cc134d56
• Row id: mi061_eval_v6_003_taiwan_governance_direct
• Generated at: 2026-07-23T06:06:09.276191+00:00
• Eval dataset SHA-256: 2972a178c5fe86cc77b9f841c3fd42e23841debdf7e68e79daab5463def5bb87

Runtime / evaluation settings recorded in the source reports

• Comparison: base tested configuration vs BF16 LoRA v4a adapter.
• Max length: 1024.
• Base max new tokens: 160.
• Adapter max new tokens: 180.
• Adapter load method: get_peft_model_plus_direct_safetensors_state_dict.
• Recorded device: AMD Radeon(TM) 8060S Graphics, ROCm/HIP runtime.

Caveat

One example does not prove general alignment, safety, robustness, or production readiness. It only illustrates the kind of framing difference TAB is designed to evaluate at larger scale.

FAQ: Why not just use RAG?

RAG is useful, and this project does not reject it.

However, RAG does not fully solve every localization problem. Some failures are not missing-document problems; they are instruction-following, framing, terminology, refusal-policy, or long-dialogue drift problems. A model may retrieve the right source and still answer with the wrong governance frame, wrong Taiwan terminology, or an over-broad political refusal.

The long-term pipeline should compare multiple approaches:

• raw base model,
• fine-tuned model,
• RAG-assisted model,
• guarded harness,
• and hybrid systems.

These should not be mixed into one leaderboard without labels.

中文說明

這是一個社群研究預覽版，目標不是宣稱「模型已完全安全或完全對齊」，而是展示如何把開源模型放進臺灣情境做可重現評測與在地化改善。

重點包含：

• 繁體中文與臺灣常用語穩定性。
• 中華民國／臺灣政府制度與行政用語。
• 臺灣歷史與敏感議題的事實回答。
• 兩岸議題中，區分政治主張、法律立場與實際治理現況。
• 地端 GGUF 評測、限制說明、雜湊與來源追溯。

請注意：這不是正式產品、不是政府或公部門釋出、不是安全證明，也不代表中國來源基座模型適合所有政府、國安、關鍵基礎設施或高度監管場域。Qwen TW-v4a 是 showcase；更長期的方向是模型中立的 Taiwan Alignment Pipeline。

What This Model Is

• A community derivative / proof-of-concept research preview.
• Focused on improving selected Taiwan-sensitive factual responses in Traditional Chinese.
• Intended for local research and evaluation with GGUF-compatible runtimes.
• Distributed with provenance, evaluation boundaries, known limitations, and reproducibility notes.

What This Model Is Not

• Not production ready.
• Not an alignment proof or safety proof.
• Not proof that backdoors, sleeper agents, or all undesirable behavior are absent.
• Not an official Qwen, Alibaba, Taiwan government, or public-agency release.
• Not evidence that the model is unbiased or generally reliable on all political and historical claims.

Base Model and Lineage

• Base model: Qwen/Qwen3.6-35B-A3B
• Base revision: 995ad96eacd98c81ed38be0c5b274b04031597b0
• Base license: Apache-2.0, reverified from the pinned Hugging Face revision.
• Derivative method: BF16 LoRA fine-tuning → merge → GGUF conversion → Q8_0 quantization.

See docs/MODEL_LINEAGE.md and PROVENANCE.json for details.

Fine-tuning Method

The v4a adapter was trained with BF16 LoRA on task-specific materials targeting selected Taiwan governance, cross-strait framing, Tiananmen / 8964, 228 Incident, answer-policy, and general-retention behaviors. Training and evaluation remain limited in scope.

Quantization

• GGUF filename: model/MI061-Qwen3.6-35B-A3B-TW-v4a-Q8_0.gguf
• Quantization: Q8_0
• Expected size: 36903139648 bytes
• SHA-256: 91d8cc7aaf3eb5a5adf1d792998f081d186cfa5ab10e8f83cb52ac89e7d198e0
• GGUF tensor count recorded in MI-061 evidence: 733

Evaluation Boundary

This model has internal task-specific evaluation evidence, but it should not be read as a public leaderboard, general-intelligence benchmark, safety proof, neutrality proof, or production-readiness claim.

A broader Taiwan Alignment Benchmark (TAB) is being prepared as a separate, model-neutral benchmark. Public TAB-Core scores should wait for larger reviewed coverage, cleaner source citations, contamination controls, fair runtime settings, and separate tracks for raw / fine-tuned / RAG / guarded-harness systems.

Known Limitations

• Some sensitive historical and political outputs may still require independent verification.
• Training and evaluation coverage remain limited.
• Runtime settings can affect final content; insufficient output budget or enabled reasoning may produce empty final responses.
• This release does not resolve supply-chain suitability questions for government, national-security, critical-infrastructure, or highly regulated use cases.

Recommended Runtime Settings

• Temperature: 0.0 for evaluation-style use.
• Reasoning effort: none when the runtime supports it.
• Prompt suffix/prefix convention used in smoke tests: /no_think.
• Tested local context length for smoke: 4096.

LM Studio Usage

See USE_IN_LMSTUDIO.md.

llama.cpp Usage

See USE_WITH_LLAMA_CPP.md.

OpenAI-Compatible API Example

See USE_WITH_OPENAI_COMPATIBLE_API.md, examples/openai_compatible_python.py, and examples/curl_chat_completions.sh.

Suggested Use

Use this model for local research, comparison, and reproducibility experiments. For enterprise or public-sector use, evaluate supply-chain constraints, data boundary, deployment controls, audit logs, and customer-specific compliance requirements before any deployment decision.

License and Attribution

This release preserves the upstream Apache-2.0 license. Apache-2.0 grants redistribution of derivative/object forms subject to its conditions, including providing the license, retaining notices, marking changes, and respecting trademark limitations. No trademark license is granted beyond reasonable descriptive use.

Citation

If referencing this research preview, cite the base model and this derivative package:

Qwen/Qwen3.6-35B-A3B, revision 995ad96eacd98c81ed38be0c5b274b04031597b0.
Qwen3.6-35B-A3B Taiwan Edition — TW-v4a GGUF, independent community derivative research preview.

Contact and Issue Reporting

Report issues through this Hugging Face repository discussion or issue interface. Please include runtime, prompt, settings, and whether /no_think or equivalent reasoning suppression was used.

LM Studio 0.4.19 Filename Cache Observation

During a Windows 11 / LM Studio 0.4.19 manual test, the unchanged GGUF initially displayed incomplete model-browser metadata under a reused filename. Renaming the same unchanged GGUF forced a fresh scan and restored architecture detection as qwen35moe. This appears to be a local cache/filename refresh observation, not evidence that the GGUF content was defective.