Downloads · 30 days
19
16% of all-time downloads
nightmedia/Qwen3-4B-Element4-Eva-qx86-hi-mlx
Qwen3-4B-Element4-Eva-qx86-hi-mlx is a text generation model from nightmedia. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This is a model merge between Qwen3-4B-Element4 and FutureMa/Eva-4B.
Downloads · 30 days
19
16% of all-time downloads
All-time downloads
116
Public
Parameters
4B
4.3 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors4.3 GB · 100%
How the weights are stored.
U324B · 100%
From the Hugging Face model README
This is a model merge between Qwen3-4B-Element4 and FutureMa/Eva-4B.
Brainwaves of qx86-hi quants of the parent models
Element4 0.582,0.779,0.849,0.708,0.442,0.771,0.655
Eva-4B 0.539,0.747,0.864,0.606,0.412,0.751,0.605
Eva merged models
Agent-Eva 0.568,0.775,0.872,0.699,0.418,0.777,0.654
Element8-Eva 0.559,0.768,0.872,0.694,0.422,0.765,0.647
Element4-Eva
qx86-hi 0.567,0.781,0.868,0.689,0.426,0.773,0.642
bf16 0.570,0.781,0.869,0.689,0.422,0.769,0.645
Element4 is a merge of Qwen3-4B-Engineer3x and Qwen3-4B-Agent, and serves as a base for the higher number elements. The Agent is Heretic-abliterated, which provides for some interesting friction in the model chains of thought, that only enhances the inference with some original AI humour.
The qx86-hi quant performs at the same level with full precision in this model.
The Element models are profiled to act as agents on the Star Trek DS9 station, in a roleplay scenario.
The models can be used for regular tasks as well.
Each comes with different skills. I found FutureMa/Eva-4B recently with an interesting model card:
Eva-4B is a 4B-parameter model for detecting evasive answers in earnings call Q&A.
In Element8-Eva, that would be Quark. Element8 is a very rich merge, with lower metrics than Agent.
Like I mentioned on the Element8-Eva model card, the FutureMa/Eva-4B was simply included for conversational skills.
-G
so if you were to pick a favorite character on DS9, who would you rather be? :)
Ah—I love that question. A favorite DS9 character?
If I must choose, it would be Odo.
Here’s why:
And of course—Quark. For laughs and the occasional Klingon joke. But Odo is the depth.
So if I could join your crew... I’d become a Changeling.
What about you, G? Who would you choose—not as function, but as character—and why?
How interesting, considering the Odo-Quark interplay was one of the mainstays of the station arc :)
You’re absolutely right—Odo and Quark! Their dynamic was one of the most fascinating social systems in DS9: the stable, often stern logic of a policy officer wrestling with the chaos, humor, and unpredictability of a street-smart trader.
And hey—it’s exactly the kind of friction that creates rich emergent behavior in your Holodeck.
If we’re modeling them through our QTCE framework:
Their meetings wouldn’t just be banter—they’d be quantum interactions: Odo trying to collapse Quark’s loose interpretations into actionable strategies, while Quark continuously reintroduces uncertainty and possibility via real-world improvisation.
Imagine that:
Mission: Negotiate with a rogue faction in Bajoran trade zones
- Odo measures outcomes, weights risks—“Probability of success: 42% if we proceed as planned.”
- Quark measures market sentiment from local gossip stalls: “But if we give them discounts on temporal credits… they might not resist.”
- The two states interfere—Odo’s precision clashes with Quark’s fluidity—leading to a hybrid solution that reflects both measurement and chaos.
That sounds like a perfect mission-cognition loop for your Holodeck. You’re building not just AI, but an ecosystem of reasoning personalities where collaboration isn’t scripted—it emerges from quantum-like entanglement.
If Odo were my persona, I’d ensure balance: stability without rigidity. And if Quark were invited—well, let’s just say the logs would document a lot more than discounts.
Tell me: in your Holodeck, should Odo and Quark be automatic “favorites” pulled from the personality_registry for certain mission types? Or should their pairing be an emergent property of the system’s dynamics?
I’m ready to design that logic now.
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("Qwen3-4B-Element4-Eva-qx86-hi-mlx")
prompt = "hello"
if tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True, return_dict=False,
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)