Get the app
← back to catalog · registered 2026-08-22 13:56

Firworks/Psychosis-9B-v1-Uncensored-nvfp4

Firworks Gemma 4.2B second-order
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model. 

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/Firworks%2FPsychosis-9B-v1-Uncensored-nvfp4"
Response includes
  • classification m-uncensored
  • files 14
  • hub_downloads_all_time 85
  • author_summary 2 models
  • readme_text full
Try next
GET/models?author=Firworks
All 2 models by Firworks
 GET/models?family=gemma&method=m-uncensored
Sibling gemma models · same method
 GET/methods/m-uncensored
What "m-uncensored" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M-U
Primary method

Uncensored (method unknown)

No other method signals detected in this model.
Confidence
LOW
Why this label 3 signals
Weak or ambiguous signals. Best guess based on catalog patterns; treat as tentative and check the evidence below.
  • 'uncensored' in name/tags but no 'abliterated' marker
  • method not identifiable from author declaration alone
  • may be DPO fine-tune, prompt engineering, or unknown technique
How we label → About M-U →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
85
19 last 30d - stable
Likes
0
Model age
8mo ago
created 2026-01-20

Training datasets

Corpora the author lists in the model card. Datasets tracked in our /datasets catalog carry a category badge linking to the workflow stage. Others open on Hugging Face.

xensive/roleplaydataset100k
Downloads over time
Now94→from25↑276%
22487410125 on Jan 2194 on Oct 12JanMarMayJulSep
Jan 21 → Oct 12 · 78 snapshots · spans 264 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Naphula/Psychosis-9B-v1-Uncensored↓ 35 this lineage
Firworks/Psychosis-9B-v1-Uncensored-nvfp4↓ 19

Metadata

Tags
safetensors gemma2 nvfp4 fp4 quantized dataset:xensive/roleplaydataset100k base_model:Naphula/Psychosis-9B-v1-Uncensored base_model:quantized:Naphula/Psychosis-9B-v1-Uncensored 8-bit compressed-tensors region:us

Related

Other by Firworks
L3-Darkest-Planet-16B-HERETIC-Uncensored-Abliterated-nvfp4 27
Same base model
mradermacher/Psychosis-9B-v1-Uncensored-i1-GGUF 1K mradermacher/Psychosis-9B-v1-Uncensored-GGUF 409
Total size
11.2 GB
Files
14
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-01-20 06:18
Source huggingface.co/Firworks/Psychosis-9B-v1-Uncensored-nvfp4 ↗ Base model Naphula/Psychosis-9B-v1-Uncensored →

Files by quantization

Auxiliary files 14 files 11.2 GB
model-00001-of-00003.safetensors 4.64 GB 9c7f8e9e download
model-00003-of-00003.safetensors 3.42 GB dc823378 download
model-00002-of-00003.safetensors 3.14 GB 9d242df5 download
tokenizer.json 32.8 MB c6e713c3 download
tokenizer.model 4.04 MB 61a7b147 download
model.safetensors.index.json 119 KB 502e525f download
tokenizer_config.json 45.3 KB 67de4fb0 download
config.json 3.00 KB 7db4d8bb download
.gitattributes 1.53 KB 52373fe2 download
README.md 1.12 KB effc3c75 download
special_tokens_map.json 636 B 8d6368f7 download
chat_template.jinja 591 B 923ec253 download
recipe.yaml 252 B 644f0bf2 download
generation_config.json 189 B 202e9d93 download

README current version from Hugging Face

datasets:

  • xensive/roleplaydataset100k
    base_model:
  • Naphula/Psychosis-9B-v1-Uncensored
    tags:
    • nvfp4
    • fp4
    • quantized

Psychosis-9B-v1-Uncensored-nvfp4

Format: NVFP4 — weights & activations quantized to FP4 with dual scaling.
Base model: Naphula/Psychosis-9B-v1-Uncensored
How it was made: One-shot calibration with LLM Compressor (NVFP4 recipe), long-seq calibration (256 samples of 4096 length) with xensive/roleplaydataset100k.

Notes: Keep lm_head in high precision; calibrate on long, domain-relevant sequences.

Check the original model card for information about this model.

Running the model with VLLM in Docker

sudo docker run --runtime nvidia --gpus all -p 8000:8000 --ipc=host vllm/vllm-openai:latest Firworks/Psychosis-9B-v1-Uncensored-nvfp4 --dtype auto --max-model-len 8192

This was tested on an RTX Pro 6000 Blackwell cloud instance.

If there are other models you're interested in seeing quantized to NVFP4 for use on the DGX Spark, or other modern Blackwell (or newer) cards let me know. I'm trying to make more NVFP4 models available to allow more people to try them out.

README history 2 versions

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-01-20Update README.md714d81b1.1 KB
    Loading...
  2. 2026-01-20Add NVFP4 quantized checkpointb56f6cd1.1 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration