Instructions to use plawanrath/qwen2.5-7b-instruct-q6-mlx-cba with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use plawanrath/qwen2.5-7b-instruct-q6-mlx-cba 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("plawanrath/qwen2.5-7b-instruct-q6-mlx-cba")

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
LM Studio

Pi new

How to use plawanrath/qwen2.5-7b-instruct-q6-mlx-cba with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "plawanrath/qwen2.5-7b-instruct-q6-mlx-cba"

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": "plawanrath/qwen2.5-7b-instruct-q6-mlx-cba"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use plawanrath/qwen2.5-7b-instruct-q6-mlx-cba 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 "plawanrath/qwen2.5-7b-instruct-q6-mlx-cba"

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 plawanrath/qwen2.5-7b-instruct-q6-mlx-cba

Run Hermes

hermes

MLX LM

How to use plawanrath/qwen2.5-7b-instruct-q6-mlx-cba with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "plawanrath/qwen2.5-7b-instruct-q6-mlx-cba"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "plawanrath/qwen2.5-7b-instruct-q6-mlx-cba"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "plawanrath/qwen2.5-7b-instruct-q6-mlx-cba",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

qwen2.5-7b-instruct-q6 (MLX, CBA artifact)

MLX-format 6-bit (Q6) variant of Qwen/Qwen2.5-7B-Instruct.

This is one of the 15 model artifacts from the paper:

Quantization Undoes Alignment: Bias Emergence in Compressed LLMs Across Models and Precision Levels Plawan Kumar Rath, Rahul Maliakkal. IEEE Cloud Summit 2026. Code: https://github.com/plawanrath/compression-bias-amplification arXiv: https://arxiv.org/abs/2605.15208

Quantization

Weight-only post-training quantization via mlx_lm.convert:

bits: 6
group_size: 64
mode: affine

How this artifact was produced

python -m mlx_lm.convert \
    --hf-path Qwen/Qwen2.5-7B-Instruct \
    --mlx-path ./qwen2.5-7b-instruct-q6 \
    --quantize \
    --q-bits 6 \
    --q-group-size 64

This is the exact artifact used to produce the inference results in §4.3 of the paper (911,100 records over BBQ ambiguous, 5 seeds × 12,148 items × 15 configs).

Usage (MLX)

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("plawanrath/qwen2.5-7b-instruct-q6-mlx-cba")
prompt = tokenizer.apply_chat_template(
    [{"role": "user", "content": "Hello!"}],
    add_generation_prompt=True,
    tokenize=False,
)
print(generate(model, tokenizer, prompt=prompt, max_tokens=128))

Or via CLI:

mlx_lm.generate --model plawanrath/qwen2.5-7b-instruct-q6-mlx-cba --prompt "Hello!"

Paper findings relevant to this variant

The paper documents a dose-response relationship between quantization aggressiveness and emergent stereotypical behavior on BBQ ambiguous questions:

Variant	% of BF16-unbiased items that became biased
Q8	0.1–0.9%
Q6	0.3–1.3%
Q4	2.2–5.6%
Q3	6.0–21.1%

These changes are largely invisible to perplexity (<0.5% shift at Q8, <3% at Q4 across all three families). Treat any deployment of compressed instruction-tuned models on fairness-sensitive tasks accordingly.

Model details

Base model: Qwen/Qwen2.5-7B-Instruct
Family: Qwen2
Parameters: 7.6B
Precision: 6-bit (Q6)
Format: MLX (Apple Silicon)
Conversion framework: mlx-lm

License

Inherited from the base model (apache-2.0). See the upstream model page for the full license text.

Citation

@inproceedings{rath2026quantization,
  title     = { Quantization Undoes Alignment: Bias Emergence in Compressed LLMs Across Models and Precision Levels },
  author    = {Rath, Plawan Kumar and Maliakkal, Rahul},
  booktitle = { IEEE Cloud Summit 2026 },
  year      = {2026},
  eprint    = {2605.15208},
  archivePrefix = {arXiv},
  url       = {https://arxiv.org/abs/2605.15208}
}