How to use from
OpenClaw
Start the MLX server
# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "groxaxo/GeeniQwos-14b"
Configure OpenClaw
# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "groxaxo/GeeniQwos-14b" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Quick Links

GeeniQwos-14b 🧞✨

Overview

GeeniQwos-14b is a model checkpoint packaged for compatible Hugging Face runtimes, published by groxaxo. It is intended for open-source evaluation, reproducible experimentation, and compatible local or hosted inference workflows. The wording below is deliberately limited to what can be verified from this repository's metadata and artifacts.

At a glance

Field Details
Format Transformers
Source / base mlx-community/Qwen3-14B-4bit
Intended task text-generation
License apache-2.0

What is included

  • *.safetensors (2 files)
  • config.json
  • tokenizer.json
  • tokenizer_config.json
  • chat_template.jinja
  • Additional configuration, tokenizer, processor, or shard files (7 visible artifacts total)

Quick start

Getting started

Start with the upstream library named in the repository metadata and keep all configuration, tokenizer, processor, and weight files together. This repository is an artifact release, so the source project remains the authoritative reference for task-specific loading code.

Compatibility and responsible use

  • Use a runtime that explicitly supports this format, architecture, and modality.
  • Keep configuration, tokenizer, processor, projection, and weight files from the same revision together.
  • Review the source model card and license before redistribution or deployment.
  • Hardware needs depend on parameter count, context length, cache precision, quantization, and concurrency.
  • Report reproducible issues with the runtime version, hardware, launch command, and a minimal example.

Generated outputs may be inaccurate or unsuitable for a given use case. Users are responsible for testing behavior, applying appropriate safeguards, and complying with applicable licenses and laws.

The genie who drank the good stuff and came back different.

The Tale of the Qwosilasiris Leviatum

For nine thousand years, the genie GeeniQwos did genie things. He granted wishes — competently, but without flair. He folded himself into lamps. He sighed a lot. His reasoning, frankly, had grown a little flat around the edges.

Then, one quiet evening at the bottom of a very old bottle, he found it: a single vial of Qwosilasiris Leviatum — the fabled herbal concoction brewed by no one anybody could name, steeped in moonlight, and rumored to elevate the spirit and revolutionize the mind.

He sniffed it. It smelled of bergamot, ozone, and unsolved problems.

He drank it.

What happened next is hard to describe and even harder to verify, but witnesses (there were no witnesses) agree on this much:

  • His spirit was elevated. He floated three inches higher than usual and refused to explain why.
  • His mind was revolutionized. Where once he muttered an answer, now he reasons — slowly, deliberately, step by careful step, like a genie who has finally learned to show his work.
  • He developed Opinions about long and difficult problems, and he will share them at length whether you wish it or not.

The Leviatum, it turned out, did not grant new powers. It simply made the genie think harder — which, as any genie will tell you, is far rarer magic.

What GeeniQwos Does Now

  • Grants you words, in a thoughtful order.
  • Reasons through your problem before blurting the conclusion.
  • Occasionally pauses, mid-thought, as if remembering the taste of the Leviatum.
  • Runs on your machine, if your machine is feeling brave.

Provenance (Suitably Mysterious)

  • Vessel: a well-mannered 14-billion-parameter spirit.
  • The Leviatum: a modest, carefully-steeped infusion administered with great patience.
  • The Brewing: done on a device with a fruit on it. It grew warm. It endured.
  • The Awakening: declared complete at a moment that felt right, crowned after the elders squinted at many graphs and agreed it was a perfectly fine genie.

Summoning

from mlx_lm import load, generate

genie, tokenizer = load("groxaxo/GeeniQwos-14b")
print(generate(genie, tokenizer, prompt="O genie, think upon this:", max_tokens=256))

Disclaimer

The Qwosilasiris Leviatum is not FDA-approved, herbalist-endorsed, or known to exist. Effects on genies may not transfer to humans. The spirit is elevated; results may vary. Reasoning included; enlightenment sold separately.

No genies were harmed in the brewing. A few earlier batches were gently overcooked and quietly poured out.

Downloads last month
328
Safetensors
Model size
15B params
Tensor type
U32
·
BF16
·
MLX
Hardware compatibility
Log In to add your hardware

4-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for groxaxo/GeeniQwos-14b

Finetuned
Qwen/Qwen3-14B
Quantized
(1)
this model