Soulfate24/Spark-X2.5-4B-ASHQ1-Remix-GGUF overview
Spark X2.5 4B ASHQ1 Remix This is a GGUF quantized version of the original model. 📈 Release Benchmarks wiki.test.raw, symmetric FA auto reference | Model | Si…
Runs locally from ~3.4 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| imatrix.gguf | GGUF | GGUF | 3.4 MB | Download |
| model-ASHQ1-Compact-33pc.gguf | GGUF | GGUF | 2.53 GB | Download |
| model-ASHQ1-Fidelity-48pc.gguf | GGUF | GGUF | 3.68 GB | Download |
| model-ASHQ1-Mini-30pc.gguf | GGUF | GGUF | 2.29 GB | Download |
| model-ASHQ1-Nano-27pc.gguf | GGUF | GGUF | 2.07 GB | Download |
| model-ASHQ1-Pico-24pc.gguf | GGUF | GGUF | 1.94 GB | Download |
| model-ASHQ1-Precision-42pc.gguf | GGUF | GGUF | 3.28 GB | Download |
| model-ASHQ1-Quality-36pc.gguf | GGUF | GGUF | 2.77 GB | Download |
Model Details
| Model ID | Soulfate24/Spark-X2.5-4B-ASHQ1-Remix-GGUF |
|---|---|
| Author | Soulfate24 |
| Pipeline | text-generation |
| License | apache-2.0 |
| Base model | XHToken/Spark-X2.5-4B |
| Last modified | 2026-09-10T14:12:15.000Z |
Model README
---
license: apache-2.0
library_name: transformers
pipeline_tag: text-generation
base_model:
- XHToken/Spark-X2.5-4B
tags:
- llm
- sparkx2_5
- quantization
- gguf
- ashq1
- imatrix
---
Spark-X2.5-4B - ASHQ1-Remix
This is a GGUF quantized version of the original model.
📈 Release Benchmarks (wiki.test.raw, symmetric FA-auto reference)
| Model | Size | PPL | KLD | RMS Δp | top-p | Speed |
| :--- | ---: | ---: | ---: | ---: | ---: | ---: |
| Q8_0 (stock) | 4172 MiB | 34.4529 | 0.0075 | 2.03% | 96.4% | 2844 t/s |
| Fidelity-48pc 🥈 | 3772 MiB | 33.8560† | 0.0185 | 2.85% | 94.0% | 2133 t/s |
| Precision-42pc ⭐ | 3359 MiB | 33.2561† | 0.0213 | 3.30% | 93.5% | 1766 t/s |
| Q6_K-imx (stock) | 3223 MiB | 33.2700† | 0.0269 | 3.73% | 92.5% | 2327 t/s |
| Quality-36pc | 2840 MiB | 34.9522 | 0.0719 | 5.88% | 88.2% | 2327 t/s |
| Q5_K_M-imx (stock) | 2840 MiB | 34.9522 | 0.0719 | 5.88% | 88.2% | 2438 t/s |
| Compact-33pc | 2593 MiB | 38.5817 | 0.1688 | 8.67% | 82.0% | 1829 t/s |
| Mini-30pc | 2343 MiB | 38.9367 | 0.2012 | 9.85% | 79.9% | 2226 t/s |
| IQ4_XS-imx (stock) | 2266 MiB | 41.9845 | 0.2586 | 10.90% | 77.5% | 2560 t/s |
| Nano-27pc ✗ | 2124 MiB | 40.6742 | 0.3281 | 12.32% | 75.3% | 2438 t/s |
| Pico-24pc ✗ | 1982 MiB | 31.7680† | 0.3910 | 13.52% | 72.4% | 1766 t/s |
| IQ3_M-imx (stock) | 1949 MiB | 41.0486 | 0.4913 | 15.49% | 69.1% | 2327 t/s |
ℹ️ About ASHQ1-Remix Suite
Activation-aware GGUF quantization whose every ratio, floor, and cap traces to a measured experiment. Plain-BF16-native first; AutoRound lineage supported with explicit saturation bounds. Full seven-tier ladder validated across six model families.
🔗 Link: https://huggingface.co/Soulfate24/AutoRound-ASHQ1-Remix_Double-Quantization_Suite
Run Soulfate24/Spark-X2.5-4B-ASHQ1-Remix-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models