GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE โ†’
Model Intelligence Sheet

Soulfate24/Ornith-1.5-9B-MTP-ASHQ1-Remix-GGUF overview

Ornith 1.5 9B with MTP ASHQ1 Remix This is a GGUF quantized version of the original model. ๐Ÿ“ˆ Release Benchmarks wiki.test.raw, symmetric FA auto reference | Mโ€ฆ

transformersggufquantizationashq1imatrixtext-generationbase_model:ornith-ai/Ornith-1.5-9Bbase_model:quantized:ornith-ai/Ornith-1.5-9Blicense:mitendpoints_compatibleregion:usconversational

Runs locally from ~4.9 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
3,938
Likes
2
Pipeline
text-generation

Repository Files & Downloads

9 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
imatrix.ggufGGUFGGUF4.9 MBDownload
mmproj-ASHQ1-Balanced-72pc.ggufGGUFGGUF631.2 MBDownload
model-ASHQ1-Compact-33pc.ggufGGUFGGUF5.81 GBDownload
model-ASHQ1-Fidelity-48pc.ggufGGUFGGUF8.24 GBDownload
model-ASHQ1-Mini-30pc.ggufGGUFGGUF5.44 GBDownload
model-ASHQ1-Nano-27pc.ggufGGUFGGUF4.64 GBDownload
model-ASHQ1-Pico-24pc.ggufGGUFGGUF4.29 GBDownload
model-ASHQ1-Precision-42pc.ggufGGUFGGUF7.38 GBDownload
model-ASHQ1-Quality-36pc.ggufGGUFGGUF6.24 GBDownload

Model Details

Model IDSoulfate24/Ornith-1.5-9B-MTP-ASHQ1-Remix-GGUF
AuthorSoulfate24
Pipelinetext-generation
Licensemit
Base modelornith-ai/Ornith-1.5-9B
Last modified2026-09-10T10:31:21.000Z

Model README

---

library_name: transformers

license: mit

license_link: https://huggingface.co/ornith-ai/Ornith-1.5-9B/blob/main/LICENSE

pipeline_tag: text-generation

base_model:

- ornith-ai/Ornith-1.5-9B

tags:

- quantization

- gguf

- ashq1

- imatrix

---

Ornith-1.5-9B with MTP - ASHQ1-Remix

This is a GGUF quantized version of the original model.

๐Ÿ“ˆ Release Benchmarks (wiki.test.raw, symmetric FA-auto reference)

| Model | Size | PPL | KLD | RMS ฮ”p | top-p | Speed |

| :--- | ---: | ---: | ---: | ---: | ---: | ---: |

| Q8_0 (stock) | 9333 MiB | 9.0824 | 0.0058 | 2.26% | 97.9% | 244 t/s |

| Fidelity-48pc | 8436 MiB | 9.0227โ€  | 0.0087 | 2.58% | 97.2% | 260 t/s |

| Precision-42pc | 7554 MiB | 8.9957โ€  | 0.0094 | 2.72% | 96.8% | 324 t/s |

| Q6_K-imx (stock) | 7209 MiB | 8.9836โ€  | 0.0107 | 2.90% | 96.4% | 320 t/s |

| Quality-36pc | 6388 MiB | 8.9895โ€  | 0.0241 | 4.22% | 94.5% | 1313 t/s |

| Q5_K_M-imx (stock) | 6335 MiB | 8.6345โ€  | 0.0665 | 6.48% | 90.7% | 1219 t/s |

| Compact-33pc โญ | 5945 MiB | 9.1332 | 0.0319 | 4.88% | 93.1% | 1219 t/s |

| Mini-30pc ๐Ÿฅˆ | 5566 MiB | 9.0189โ€  | 0.0425 | 5.65% | 91.8% | 1384 t/s |

| IQ4_XS-imx (stock) | 5080 MiB | 9.2758 | 0.0538 | 6.29% | 90.7% | 1506 t/s |

| Nano-27pc | 4750 MiB | 9.5968 | 0.0802 | 7.57% | 87.9% | 1249 t/s |

| Pico-24pc | 4389 MiB | 9.5871 | 0.1202 | 9.25% | 85.1% | 1384 t/s |

| IQ3_M-imx (stock) | 4313 MiB | 9.5973 | 0.1421 | 10.46% | 84.3% | 1384 t/s |

โ„น๏ธ 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.

๐Ÿ”— Link: https://huggingface.co/Soulfate24/AutoRound-ASHQ1-Remix_Double-Quantization_Suite

Run Soulfate24/Ornith-1.5-9B-MTP-ASHQ1-Remix-GGUF with guIDE

Download guIDE โ€” the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE โ†’ ยท Browse 524k+ models ยท Compare models

Source: Hugging Face ยท Compare models