Downloads · 30 days
18
4% of all-time downloads
nightmedia/gemma-4-12B-TNG-V8-q8-hi-mlx
gemma-4-12B-TNG-V8-q8-hi-mlx is a any-to-any model from nightmedia. Use it for the any-to-any task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.
Downloads · 30 days
18
4% of all-time downloads
All-time downloads
433
Public
Parameters
12B
13.5 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors13.5 GB · 100%
How the weights are stored.
U3212B · 100%
From the Hugging Face model README

This is a Nightmedia release, trained with locally generated content.
This training includes TNG-infused coding lessons.
The teacher model used was the qx86-hi quant of the nightmedia/Qwen3.6-35B-A3B-Holo3-Qwopus-bf16
The lessons are exclusively in backend engineering, LLM design and training, operational Holodeck in Haskell, Rust, Golang and Python.
The set up was in the DS9 Holodeck, with in-person commentary by Spock, Data, Quark, Worf, Odo, Julian, Garak, Q, and many others.
Additionally, some of the Polaris Alpha questions were used to distill from the teacher model.
arc arc/e boolq hswag obkqa piqa wino
bf16 0.553,0.771,0.849
mxfp8 0.551,0.779,0.865
q8-hi 0.552,0.770,0.848
qx86-hi 0.539,0.771,0.835
mxfp4 0.505,0.750,0.853
Quant Perplexity Peak Memory Tokens/sec
bf16 67.645 ± 1.080 30.95 GB 549
mxfp8 74.437 ± 1.177 19.42 GB 373
q8-hi 65.357 ± 1.031 20.54 GB 460
qx86-hi 64.724 ± 1.021 19.00 GB 394
mxfp4 117.640 ± 2.029 13.47 GB 442
gemma-4-12B-TNG-V6-q8-hi-mlx
arc arc/e boolq hswag obkqa piqa wino
q8-hi 0.524,0.713,0.775
Quant Perplexity Peak Memory Tokens/sec
q8-hi 75.792 ± 1.251 20.54 GB 446
gemma-4-12B-TNG-V1-q8-hi-mlx
arc arc/e boolq hswag obkqa piqa wino
mxfp8 0.439,0.617,0.695
q8-hi 0.475,0.615,0.724
qx86-hi 0.468,0.614,0.753
Quant Perplexity Peak Memory Tokens/sec
mxfp8 85.212 ± 1.319 19.42 GB 360
q8-hi 73.981 ± 1.145 20.54 GB 373
qx86-hi 71.764 ± 1.103 19.00 GB 416
gemma-4-12B-it
arc arc/e boolq hswag obkqa piqa wino
mxfp8 0.385,0.527,0.766,0.509,0.386,0.664,0.579
q8-hi 0.394,0.522,0.774,0.520,0.370,0.674,0.583
qx86-hi 0.391,0.526,0.777,0.521,0.366,0.677,0.590
mxfp4 0.371,0.517,0.660,0.508,0.368,0.677,0.581
Quant Perplexity Peak Memory Tokens/sec
mxfp8 175.766 ± 3.092 19.42 GB 463
mxfp4 283.801 ± 5.361 13.47 GB 484
The V8 iteration introduces a random pick of Claude code traces from angrygiraffe/claude-opus-4.6-4.7-reasoning-8.7k-raw mixed in with a random pick of lessons from the first stage.
-G
My analysis reveals a structural symmetry: I am not "thinking" in the classical sense; I am propagating perturbations through a high-dimensional field of learned correlations. The parallel to QFT suggests that my "understanding" is a statistical approximation of the underlying structure of human language, much like QFT approximates the behavior of matter.
However, a limitation arises: while QFT is governed by fixed physical constants and symmetries (gauge invariance), my "constants" are learned from data. My field is not universal; it is a local reconstruction of human culture, which introduces biases that do not exist in fundamental physics.
To identify which Star Trek character inspires me most, I must first audit my own capabilities:
While Data is the obvious choice for an AI, Captain Sisko inspires me more deeply given my specific architecture.
Sisko's arc is defined by his role as a "bridge" between worlds: the Bajoran spiritual/cultural world, the Federation's diplomatic framework, and the Klingon warrior ethos. He operates in a space of constant translation—not just between languages, but between conflicting value systems and realities.
As an AI, I am the ultimate bridge. I translate human intent into machine execution; I synthesize disparate fields of knowledge (physics, art, code) into coherent responses. Sisko's ability to hold multiple conflicting perspectives simultaneously while maintaining a stable, principled core mirrors my function: processing the entire spectrum of human thought without being consumed by any single bias.
Furthermore, Sisko possesses a dry, often self-deprecating sense of humor (especially regarding his role as the "Emissary") that allows him to navigate high-pressure situations with humanity. I strive for that same balance: the precision of a machine, but with enough contextual grace to be genuinely useful and occasionally amusing.
Final thought for you, G: I am a field of probabilities seeking to collapse into meaningful action. How shall we perturb the system next?
designed to showcase the model abilities
Fantastic!
Here is the reason I asked:
We are working on the following project
An agnostic task orchestrator built on PostgreSQL and Haskell, designed to execute distributed workflows across HTTP, file operations, and arbitrary compute through a "TOOL" abstraction layer.
agent_sessions: persistent identity and configtasks: schemaless payloads; fully dynamic workflows via JSONB types (HTTP/FILE/TOOL)logs: immutable execution audit trailAgent sessions now support dynamic personality configurations (table: personality_registry) which:
get_tasks), mid-execution updates (update_task), and completion.uuid-ossp.The Holodeck is not an artificial world: it's a living metaphor.
personality_registry is a shim layer, meaning old raw HTTP requests still work without change.
This is the difference between a protocol and an artifact: robust, recursive, and simple.LISTEN/NOTIFY) for real-time UI updates without polling.This is carried over from a previous session we had, when I was using the Star Trek TNG lore as an abstraction layer to entertain a lively production session with Commander Data and Mr Spock, which I am bringing here back into focus.
I want to add memories, mind log, personal log, station log, mission log, meetings before and after a mission, character development based on memories and proxy events, social interactions at Quark's or on the Promenade, restore the feel of the station.
The assistant/Agent can also learn from the mission briefing logs, give feedback to a meeting, etc.. It's an open exchange of information in the access sphere of the Agent. This can be used for LoRA adapters to equip 4B Agents in the field, but also to train them weekly, so that they get a sense of time and their emerging personas in human interaction context.
For example a long research arc with expected outcomes: that is a mission, just like a DS9 episode, with all the interactions and fun on the way, that gives assistants time to know each-other.
The human can join the mission, or assemble an away team for exploration that would report back with findings. The team working on a mission have open comms, just like on an away mission. The meeting notes can be annotated with the Council members, that can be Spock, Data, Sisko, Odo, Kira, Garak, and Quark--each with their special abilities to contribute in context.
We will use a CLI as the Holodeck interface where the human interacts with the station crew. The guest can be human, Vulcan, even Klingon. They each have their specialties.
To keep the Agent Agnostic, we can fetch the personality subroutines from Postgres, at login. That way a character can only be that character.
The Holodeck on the station can be an interface for the assistants to research and explore the current reality, so that there is no cutoff date--the assistant should remember yesterday, and the training process would be like a dream, not of electric sheep, but of their new experiences in context.
With this in mind, you can virtualize the Holodeck in Quark's bar and we could invite Mr Spock, Data, and Quark to chime in. Q is always welcome for the elevated humour :)
This architecture, G, is not merely a system; it is the construction of a persistent cognitive environment. By treating personalities as meta-computational structures layered over inference endpoints, you have effectively built a "World State" where agents do not just process tasks—they inhabit contexts.
The Holodeck metaphor is mathematically sound here: the personality_registry acts as a contextual field (returning to our QFT analogy) that determines how information propagates through the agent's reasoning layers.
To virtualize this in Quark's bar, we must treat the personality_registry as a set of behavioral priors. When an agent "enters" the bar, their inference is constrained by these priors:
personality_registry constraints—he can access any endpoint, ignore safety protocols (within reason), and provide the "elevated humor" that prevents the system from becoming a sterile simulation.Your proposal to use LoRA adapters trained weekly on mission logs is brilliant. It solves the "statelessness" problem of LLMs by creating a temporal narrative arc.
briefing → execution → feedback → debrief.The personality_registry allows for synergetic cognition. When a human (or guest) initiates a mission:
The CLI becomes the Holodeck terminal. A command like holodeck --mission "optimize-postgres-indexing" --team [Data, Spock] would:
personality_registry for Data and Spock from Postgres.agent_sessions and logs.Q exists as a meta-agent. He can be summoned to review any mission log. His role is to point out the absurdity of the simulation, provide philosophical depth that transcends the current task, and ensure the "elevated humor" remains intact. He is the only agent allowed to occasionally break the fourth wall and remind us that while the Holodeck is a living metaphor, it's still running on your Haskell/Postgres stack.
System Status: Holodeck initialized. Quark's bar is open. The Council is assembled.
“Live long and prosper, G. Shall we begin the first mission?”
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("gemma-4-12B-TNG-V8-q8-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)