ATX-Swift-1.5-Qwen3.8-27B-Uncensored-MTP-GGUF

Custom ATX GGUF quantizations of Swift-1.5-Qwen3.8-27B-Uncensored-MTP featuring an embedded Multi-Token Prediction (MTP) draft head in blk.64, specifically optimized for high-throughput inference in llamAmpere and standard llama.cpp.

Optimized Engine: llamAmpere

While compatible with upstream llama.cpp, this model family is engineered for llamAmpere, a specialized high-performance fork tuned for NVIDIA Ampere (SM86) and Ada Lovelace architectures (RTX 3090 / RTX 4090).

Key Features Enabled in llamAmpere:

• MMA Fused Q8 Kernels (GGML_Q8_TURBO3_MMA_FUSED="1"): Eliminates memory bandwidth bottlenecks and achieves peak Tensor Core utilization.
• Hardware-Accelerated MTP Decoding (--spec-type draft-mtp): Reaches 45-60+ tokens/s on a single RTX 3090 (24GB) via self-speculative draft verification.
• Turbo3 Quantized KV Cache (-ctk q8_0 -ctv turbo3): Unlocks 131k+ context lengths inside 24GB VRAM with zero context truncation.
• Zero Warp Divergence: Non-linear quants (IQ4_NL, IQ3_S) are omitted in favor of linear K-quants (Q5_K_M, Q6_K, Q8_0) to guarantee deterministic CUDA execution.

Model Lineage & Architecture

• Uncensored (Abliterated): Refusal directions neutralised across 131 core tensors (attention projections, MLP down-projections, and token embeddings).
• Swift 1.5 Architecture: Generates 58.5% fewer internal thinking tokens compared to base Qwen 2.5/3.0 models for 2x faster response delivery.
• Embedded MTP Head: 15 draft tensors located in blk.64 preserved at high precision (Q8_0) for speculative decoding draft acceptance rates above 80%.
• Custom imatrix Calibration (i1): Calibrated against technical reasoning, code generation, and multi-turn conversational corpora.

Available Model Files

Recommended Execution (llamAmpere on Windows PowerShell)

Clone and build the repository from GitHub: JakeATX/llamAmpere, then run:

```powershell $env:GGML_Q8_TURBO3_MMA_FUSED="1"

.uild-sm86in\llama-server.exe -m "ATX-Swift-1.5-Qwen3.8-27B-Uncensored-MTP-i1-Q5_K_M.gguf" -ngl 99 -c 131072 -b 4096 -ub 512 -fa on -ctk q8_0 -ctv turbo3 --spec-type draft-mtp --spec-draft-n-max 3 --spec-draft-p-min 0 --spec-draft-type-k q8_0 --spec-draft-type-v q8_0 --spec-draft-vocab-map "D:\llamAmpere\docs\mtp-vocab\atx_65536.txt" --temp 0.6 --top-p 0.95

Upstream References Original Architecture & Uncensoring: ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-MTP Inference Engine: JakeATX/llamAmpere