license: apache-2.0
base_model: JonathanColetti/Qwen3.8-27B-Uncensored
library_name: gguf
tags:

• text-generation
• gguf
• quantizer
• autoround
• architecture-aware
• mamba
• ssm
• multi-token-prediction
• mtp
• uncensored
• 27b

Qwen3.8-27B-Uncensored-YMQ-MTP-GGUF

Source Model: JonathanColetti/Qwen3.8-27B-Uncensored

⚖️ An Architecture-Aware, AutoRound-Inspired Mixed Precision Layout

This repository features advanced, custom architecture-aware quantizations of Qwen3.8-27B (Uncensored) processed directly from official raw BF16 source files using the custom YMQ-Compiler (v2.0) log-space framework.

These builds natively support parallel multi-token prediction (MTP) speculation engines and utilize high-context optimization parameters tailored for demanding code development API execution environments (such as RooCode/Aider).

📊 Quantization Preset Tier Details

🔄 Notice: The XS preset has been re-quantized and re-uploaded to fix a precision layout issue on the fallback layers. This hotfix resolves the API calling failures and context repetition bugs during heavy agent workloads.

📉 Perplexity Evaluation Metrics (WikiText-2)

The following metrics demonstrate the mathematical quality preservation of the YMQ-Compiler log-space cluster analysis compared to standard linear quantization layouts. Tested natively via llama-perplexity over a 4096 context window using the official WikiText-2 test corpus.

💡 The Uncensored Performance Breakthrough

Notice that the Uncensored M preset achieves an elite score of 6.8176, outperforming even the original base model's score (6.8413). This occurs because removing the artificial refusal safety layers allows the model's underlying Attention and Mamba SSM weights to predict text paths with absolute, unrestricted mathematical clarity.

By pairing JonathanColetti's pristine abliteration weights with the YMQ-Compiler's log-space gate insulation, this preset matches the raw reasoning power of the massive 19GB XL file while clawing back a clean 5 Gigabytes of VRAM overhead cache space for local RooCode/Aider coding loops!

⚖️ YMQ vs. Uniform Quantization (The AutoRound Philosophy)

Standard quantization pipelines apply a blunt, uniform bit-depth across every single layer in a model. This wastes valuable VRAM on silent background layers while starving critical logic anchors of necessary precision.

The YMQ-Compiler implements a philosophy similar to advanced weight-tuning frameworks like Intel's AutoRound:

• Targeted Bit Allocation: It strips bits away from low-leverage background tensors and automatically re-allocates that saved VRAM budget straight into full high-fidelity shields for the model's highest cognitive spikes and boundary pathways.
• Instant Optimization: Instead of running heavy, days-long optimization training loops, YMQ achieves a highly accurate mixed-precision layout instantly by analyzing layer importance metrics in log-space.

The result is a custom mixed-precision portfolio that matches the low perplexity and high context stability of premium optimized quants (like AutoRound), while maintaining an ultra-lightweight, high-speed single-GPU cache footprint.

🛠️ The YMQ Compilation Architecture

Standard quantization pipelines treat network tensors like a flat dataset, applying destructive blanket low-bit compression to delicate tracking networks. The YMQ-Compiler solves high-context logic decay by parsing model files dynamically via an automated, multi-tiered protection matrix:

1. Log-Space Gap Detection Clustering: Instead of flat percentage thresholds, the engine computes statistical cluster variances in log-space, successfully isolating intermediate logical reasoning spikes and elevating them to stable non-linear 4-bit (IQ4_XS) formats, while compressing idle fact-storage layers to aggressive 2-bit baselines.
2. Fading Boundary Tapering: Recognizes the extreme fragility of initial token entry data vectors, forcing an input wave cushion (L00=IQ4_NL → L01=IQ4_XS → L02=IQ3_XXS) that gradually stabilizes parameters before hitting the fallback pools.
3. Dedicated Gate Insulation: Hard-shields volatile parallel Transformer Multi-Head Attention and Mamba Linear State Space Model (SSM) routing paths, keeping context tracking perfectly noise-free.
4. Asymmetric Vocabulary Shielding: Fixes tied-weight boundary errors by mapping the final logit classification exit heads to robust configurations to completely eliminate formatting loops and API tag leakage under deep contexts.
5. Native Next-N Speculative Stripping: Processed with advanced pre-tokenizer stripping to ensure zero index offset drift or layer-shifting risks across hybrid configurations.

🚀 Recommended Runtime Parameters (llama.cpp / llama-server)

$./llama-server -m models/Qwen3.8-27B-Uncensored-YMQ-M.gguf -ctk q8_0 -ctv q4_0 --ctx-size 245760 --mmproj models/Qwen3.8-27B-Uncensored-vision-Q6_K.gguf \
  --spec-type draft-mtp --spec-draft-n-max 2 --timeout 36000 --checkpoint-min-step 2048 --ctx-checkpoints 4 \
  --n-predict -1 --temp 0.6 --top-p 0.95 --top-k 20 --repeat-penalty 1.05 --jinja -fa

🖼️ Vision Projection (--mmproj)

For multimodal vision support, pair these builds with one of the following projection files:

Pass via --mmproj <path-to-file> in your llama-server invocation (see example above).

☕ Support & Future R&D

If the YMQ-Compiler builds saved your context window from collapsing or optimized your active development cycle speeds, consider buying a coffee to fund further low-level optimization research. Your support keeps the server nodes baking future model scales!

👉 Support ZeroDigest Research on ko-fi

📦 Source Framework & Automation Code

The compiler pipeline automation engine, setup thresholds, and structural mapping rules are open-source. To view the implementation details or compile your own custom models natively using this profile layout, visit the official development hub:

👉 GitHub: ZeroDigest / YMQ-Compiler