Ornith-1.5-122B-GGUF by unsloth

Ornith-1.5-122B-GGUF by unsloth

Runs the Ornith-1.5-122B text model on your own machine, as GGUF files made by unsloth. A 122B mixture-of-experts model aimed at coding and agent work on a single GPU.

Made by unsloth on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: This is a GGUF copy of Ornith-1.5-122B, a text generation model from ornith-ai, packed by unsloth for local use. The model is a 122B mixture-of-experts design meant to run on a single GPU. At 4-bit the files come to about 62.3 GB, and the license is MIT.

Who it suits: Developers and teams who want a large open model on their own hardware, for coding and agent-style tasks, without sending prompts to a hosted service.

Industry | Who uses it | How they use it | Example
Software | Developers | Run a large local model for coding help | A terminal assistant that works offline
Research | AI labs and students | Compare an open model against others on coding benchmarks | A side-by-side test on a task set
IT operations | Teams with private code | Keep source and prompts on their own servers | An internal code reviewer

If your computer has little memory or no strong GPU, you can skip it, since the 4-bit files alone are about 62.3 GB.

Where to use it:

• Windows, macOS or Linux with Python, for the download
• Unsloth Desktop, which the card names as the way to run it
• Other programs that load GGUF files

Use case | Who | What they get back
Coding assistance | Developers | Code and fixes written by a local model
Agent work | Builders of agent tools | A model trained through self-improvement on tasks and strategies
Benchmark comparison | Researchers | Scores to set beside other open models

How to set it up:

• Python, to run the Hugging Face download tool
• About 62.3 GB of free disk space for the 4-bit files
• Hardware able to hold a model that size, ideally one large GPU

1. Install the Hugging Face hub tools with pip.
2. Download the GGUF files with the hf command.
3. Open the files in Unsloth Desktop or another GGUF-capable program.

pip install -U huggingface_hub
hf download unsloth/Ornith-1.5-122B-GGUF

How to use it: The card points to Unsloth Desktop for running the model. Load the downloaded GGUF files there, then type your prompt in its chat window as you would with any text model.

Example: You load the files in Unsloth Desktop and type "Write a shell script that renames every file in a folder to lowercase". The model replies with a script and a short explanation of what each line does.

For newcomers and experts: Start by running the two download commands and opening the files in Unsloth Desktop. Try one small coding question first, since a model this size is slow on modest hardware.

Version 1.5 follows an earlier 1.0 release that started from Qwen3.5 and Gemma4 models. Its training loop generates its own tasks, designs its own solving setups and learns from rollouts through reinforcement learning. A quantized Ornith-1.5-122B-Mobile variant exists for phones.

Download the GGUF files
Open them in Unsloth Desktop
Type a prompt
The model generates a reply

Good to know:

• License: MIT
• Cost to run: none named, but you need your own hardware and about 62.3 GB of disk at 4-bit
• Data it reaches: downloads files from Hugging Face, then runs where you load it
• Not for: machines that cannot hold about 62.3 GB of model files

Title: TerminalBench 2.1 scores in the card's table
Ornith-1.5-122B: 81.2
Qwen3.8-Flash-Next: 83.4
Qwen3.8 27B dense: 71.9
Nemotron 3.5 Super VL: 66.7

On Hugging Face, 13 November 2026

• Task: text generation
• Parameters: 125B
• Weights: about 62.3 GB at 4-bit (estimate)
• Formats: GGUF
• Downloads in the last 30 days: 680,483
• Likes: 101
• License: mit
• Based on: ornith-ai/Ornith-1.5-122B
• Tags: gguf, ornith, qwen3_5_moe, unsloth, text-generation, conversational

Links

• Model: https://huggingface.co/unsloth/Ornith-1.5-122B-GGUF
• Files: https://huggingface.co/unsloth/Ornith-1.5-122B-GGUF/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download unsloth/Ornith-1.5-122B-GGUF

Then follow the usage in its model card. Nothing in this kit runs by itself.