TeichAI/Qwen3.6-27B-Fable-5-Experimental-GGUF overview
Qwen 3.6 27B Claude Fable 5 Experimental <div style="border left: 4px solid D97706; background: rgba 217, 119, 6, 0.08 ; padding: 14px 16px; border radius: 8px…
Runs locally from ~884.6 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| Qwen3.6-27B-Fable-5-Distill.bf16.gguf | GGUF | GGUF | 50.90 GB | Download |
| Qwen3.6-27B-Fable-5-Distill.iq4_nl.gguf | GGUF | GGUF | 15.59 GB | Download |
| Qwen3.6-27B-Fable-5-Distill.q3_k_m.gguf | GGUF | GGUF | 13.18 GB | Download |
| Qwen3.6-27B-Fable-5-Distill.q3_k_s.gguf | GGUF | GGUF | 12.04 GB | Download |
| Qwen3.6-27B-Fable-5-Distill.q4_k_m.gguf | GGUF | GGUF | 16.20 GB | Download |
| Qwen3.6-27B-Fable-5-Distill.q5_k_m.gguf | GGUF | GGUF | 18.70 GB | Download |
| Qwen3.6-27B-Fable-5-Distill.q6_k.gguf | GGUF | GGUF | 21.36 GB | Download |
| Qwen3.6-27B-Fable-5-Distill.q8_0.gguf | GGUF | GGUF | 27.42 GB | Download |
| mmproj-BF16.gguf | GGUF | BF16 | 888.0 MB | Download |
| mmproj-F16.gguf | GGUF | F16 | 884.6 MB | Download |
| mmproj-F32.gguf | GGUF | F32 | 1.72 GB | Download |
Model Details
Model README
---
base_model: TeichAI/Qwen3.6-27B-Fable-5-Experimental
tags:
- gguf
- llama.cpp
- unsloth
- qwen3_5
license: apache-2.0
language:
- en
datasets:
- armand0e/claude-fable-5-claude-code
---
Qwen 3.6 27B - Claude Fable 5 (Experimental)
<div style="border-left: 4px solid #D97706; background: rgba(217, 119, 6, 0.08); padding: 14px 16px; border-radius: 8px; margin: 16px 0;">
<strong style="color: #D97706;">Update · 6/22/2026</strong><br>
GGUFs re-done with MTP built in.<br>
<strong>llama.cpp MTP args:</strong>
<code>--spec-type draft-mtp --spec-draft-n-max 3</code>
</div>
Heres Qwen3.6 slightly over-trained on a very small dataset of fable 5 traces.
Had to use aggressive settings for the style transfer here due to the data contraints. Overall the model seems to be a better planner now and better at 3D modeling in three.js, making small games, and ML engineering.
Either way give it a shot, it really does look, talk, and plan like fable... but like all fine-tunes it comes with it's limitations.
Hopefully we can get some more well rounded data from the rest of the community to a do a much less aggressive tune on a larger dataset.
Reasoning was left untouched
Benchmarks
As always big thanks to @nightmedia for the speedy benchmarks.
arc arc/e boolq
Qwen3.6-27B-Fable-5-Experimental 0.650 0.813 0.909
Qwen3.6-27B 0.637 0.798 0.911
Not really a benchmark but here's a procedurally generated duck that it zero-shotted lol (quant: q3_k_s)
---
The data for this model was easily extracted, formatted, and masked for training with Teich <img src="https://cdn-avatars.huggingface.co/v1/production/uploads/6837935ac3b7ffe0d2559ce9/-AxyvV4wfUY8uo87kNKkK.png" width="20" height="20" style="display: inline-block; vertical-align: middle; margin: 0 3px;">
This model was trained 2x faster with Unsloth and Huggingface's TRL library.
Run TeichAI/Qwen3.6-27B-Fable-5-Experimental-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models