---
base_model:
- ornith-ai/Ornith-1.5-397B
base_model_relation: quantized
library_name: llama.cpp
tags:
- gguf
- imatrix
- quantization
- mtp
---

# Ornith-1.5-397B MTP GGUF
- vision: yes
- MTP: yes!
- best quality / byte: yes. pretty sure.
- bartowski imatrix data
- Unsloth-style mixed quants


![PPL vs size](quality-vs-size-ppl.png)

![Top-token agreement vs size](quality-vs-size-topp.png)

[`results.tsv`](results.tsv).

## Sources

- Base model: [ornith-ai/Ornith-1.5-397B-GGUF](https://huggingface.co/ornith-ai/Ornith-1.5-397B-GGUF) — `Ornith-1.5-397B-Q8_0.gguf`
- Calibration corpus: [bartowski/Ornith-1.5-397B-GGUF](https://huggingface.co/bartowski/Ornith-1.5-397B-GGUF) — `Ornith-1.5-397B-calibration-v6.txt` (mirrored here)
- Tensor-protection recipe lineage: [unsloth/Ornith-1.0-397B-GGUF](https://huggingface.co/unsloth/Ornith-1.0-397B-GGUF) UD tables (same `qwen35moe` arch), read from shard headers