Instructions to use DreamFoundries/Ornith-1.0-35B-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DreamFoundries/Ornith-1.0-35B-8bit with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("DreamFoundries/Ornith-1.0-35B-8bit")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use DreamFoundries/Ornith-1.0-35B-8bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "DreamFoundries/Ornith-1.0-35B-8bit"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "DreamFoundries/Ornith-1.0-35B-8bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use DreamFoundries/Ornith-1.0-35B-8bit with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "DreamFoundries/Ornith-1.0-35B-8bit"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default DreamFoundries/Ornith-1.0-35B-8bit

Run Hermes

hermes

MLX LM

How to use DreamFoundries/Ornith-1.0-35B-8bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "DreamFoundries/Ornith-1.0-35B-8bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "DreamFoundries/Ornith-1.0-35B-8bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "DreamFoundries/Ornith-1.0-35B-8bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Ornith-1.0-35B-8bit

This repository contains an MLX-LM conversion of deepreinforce-ai/Ornith-1.0-35B.

Conversion Details

Original model: deepreinforce-ai/Ornith-1.0-35B
Model size: 35B
MLX model type: qwen3_5_moe
Quantization: MLX-LM affine quantization
Bits: 8-bit
Group size: 64
Local MLX folder size: 34.32 GiB
Local safetensors weight size: 34.30 GiB

This conversion used a small MLX weight-name adaptation for the MoE expert tensors before quantization.

Usage

mlx_lm.generate --model DreamFoundries/Ornith-1.0-35B-8bit --prompt "Hello"

Benchmarks

No comparative benchmarks have been run for this conversion. The uploaded files were checked locally by loading the model with mlx_lm.generate and generating a short sample, but no quality, speed, memory, or benchmark comparisons against the original Hugging Face weights or other quantizations are provided.

License

The original model is released under the MIT license. See the original repository for the upstream model card, license, and usage notes: deepreinforce-ai/Ornith-1.0-35B.

Downloads last month: 2

Safetensors

Model size

35B params

Tensor type

BF16

U32

MLX

Hardware compatibility

8-bit

Model tree for DreamFoundries/Ornith-1.0-35B-8bit

Base model

deepreinforce-ai/Ornith-1.0-35B

Quantized

(81)

this model

Collection including DreamFoundries/Ornith-1.0-35B-8bit

Ornith 1.0 MLX Quantizations

Collection

MLX-LM affine quantizations for Ornith 1.0 models; see each model card for details. • 6 items • Updated about 12 hours ago