← back to catalog · registered 2026-11-16 12:03

LTM-Lab/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15

LTM-Lab 27B multimodal 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/LTM-Lab%2FQwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15"
Response includes
  • classification m3
  • files 30
  • author_summary 17 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
Why this label 2 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • 'heretic' in model name (Heretic-produced)
  • Heretic uses layer-wise optimization (M3) with underlying direction removal (M1)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
Downloads · 30-day
124
Likes
0
Descendants
3
in 2 direct forks
Model age
5d ago
created 2026-11-16

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.

Downloads over time
Now 124 → from 0 ↑ 0%
04591136 0 on Nov 16 124 on Nov 22 Nov
Nov 16 → Nov 22 · 7 snapshots · spans 6 days

Genealogy 2 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.

Variants by this author 2 formats · 248 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
apache-2.0
Languages
ar zh en fr de hi id it ja ko pl pt ru es th vi
Tags
transformers safetensors qwen3_5 image-text-to-text v15-episodic-memory sglang fp8 heretic qwen3_8 qwen3_6 uncensored finetune

Related

Total size
51.7 GB
Files
30
Quantizations
1
Registered
2026-11-16 12:03
Last updated on HF
2026-11-18 10:23

Files by quantization

Auxiliary files 30 files 51.8 GB
model-00003-of-00012.safetensors 4.62 GB 1179dd73 download
model-00005-of-00012.safetensors 4.62 GB 57a59c5a download
model-00007-of-00012.safetensors 4.62 GB 89112ff2 download
model-00009-of-00012.safetensors 4.62 GB c1f2ae5a download
model-00008-of-00012.safetensors 4.59 GB b3bba7a0 download
model-00010-of-00012.safetensors 4.59 GB 0a5772c3 download
model-00004-of-00012.safetensors 4.59 GB e78a749a download
model-00006-of-00012.safetensors 4.57 GB ed27daac download
model-00011-of-00012.safetensors 4.57 GB a0c774ce download
model-00002-of-00012.safetensors 4.51 GB a5e8e851 download
model-00012-of-00012.safetensors 3.34 GB 396464aa download
model-00001-of-00012.safetensors 2.37 GB c098f75a download
model-mtp-restored.safetensors 100 MB fbea2586 download
tokenizer.json 12.2 MB 0997f410 download
vocab.json 6.65 MB c36b3f76 download
liftoff-cooking.gif 1.35 MB cb69143e download
modeling_qwen_dense_v15.py 261 KB bd34965d download
model.safetensors.index.json 111 KB 490712ba download
chat_template-tturbo-v2.jinja 38.8 KB 46a5578a download
README.md 28.3 KB 5263dd85 download
chat_template.jinja 17.0 KB b26fe3b4 download
chat_template_fixed.jinja 14.8 KB da09c9ea download
configuration_qwen_dense_v15.py 10.6 KB 10c57b10 download
tokenizer_config.json 8.78 KB a6492ceb download
config.json 4.53 KB a4936079 download
.gitattributes 1.62 KB bdef83e7 download
processor_config.json 1.22 KB fea06ab6 download
preprocessor_config.json 410 B 82b2c0d3 download
video_preprocessor_config.json 405 B d48791eb download
generation_config.json 226 B 3ba7689d download

README current version from Hugging Face


license: apache-2.0
pipeline_tag: image-text-to-text
library_name: transformers
tags:

  • v15-episodic-memory
  • sglang
  • fp8
  • heretic
  • qwen3_8
  • qwen3_6
  • uncensored
  • finetune
  • Cold Fusion
  • GAIN Training
  • Multi-stage tuning
  • all use cases
  • unsloth
    datasets:
  • DavidAU/Polar-STRICT-Datasets
  • DavidAU/F451-STRICT-Datasets
  • DavidAU/THE-DECKARD-Datasets
    language:
  • ar
  • zh
  • en
  • fr
  • de
  • hi
  • id
  • it
  • ja
  • ko
  • pl
  • pt
  • ru
  • es
  • th
  • vi
    base_model:
  • DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored

Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15

繁體中文摘要:這是 DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored 的 V15 轉換版。權重與原模型逐位元相同,多的是記憶的設定與兩個程式檔。用 Docker 鏡像 lovewin10/sglang-v15:nightly-dev-cu13-20260924-ffac53d7-v15(修改過的 SGLang)服務時,模型帶有跨請求保存的情節記憶:在一段對話裡說過的事,之後在另一段全新的對話裡問得到。V15 與 V14 相比拿掉了世界模型(記憶不再擬合任何東西,完全不需要訓練),多了睡眠:伺服器閒置時,用被服務的模型自己整理記憶庫,重播新的記憶、合併重複的、請模型判斷新的事實是否取代了舊的、在講同一個人事物的筆記之間搭橋;夢與判斷都不寫入記憶,也不刪除任何記憶。CUDA graph、前綴快取與推測解碼照常開著。實測結果與已知的限制寫在下面。

DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored converted to the V15 episodic memory architecture (Qwen3_5V15ForConditionalGeneration).

  • The weights are the original's, byte for byte. Nothing was trained, merged or quantized here, and the memory trains nothing either.

  • Added: the architecture name, an auto_map and the memory options (qwen_dense_v15_memory_config) in config.json, and two code files (configuration_qwen_dense_v15.py, modeling_qwen_dense_v15.py) that hold the memory.

  • Served by the Docker image lovewin10/sglang-v15:nightly-dev-cu13-20260924-ffac53d7-v15, a modified SGLang, the model has a memory that persists across requests: what was said in one chat is answered in a later, separate chat. The store is on disk and survives restarts.

  • Against V14 (LTM-Lab/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v14): the memory's world model is gone (small predictors fitted to the store, the only training V14 did; they did not measurably help), and while the server is idle the memory sleeps (below).

  • model-mtp-restored.safetensors (from the source repository) holds the MTP head the speculative decoding of the command below drafts with.

  • chat_template_fixed.jinja is an additional chat template from the local copy the conversion was made from; the command below uses chat_template.jinja, the source repository's.

  • An FP8 quantization is LTM-Lab/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15-FP8: 28.5 GB of weights instead of 51, faster decoding, the V15 image measured on it.

How the memory works

While a prompt is read, the keys and values its attention layers compute are cut into events of up to 128 tokens and written to a store on disk, each with a low-rank engram of the state it leaves in the model's recurrent layers. A later prompt is read once to see what it asks for; the stored events whose keys would take its attention are recalled by one vote for all layers and heads; their engrams are composed into the request's recurrent state and the prompt is read again with the recalled keys and values merged into the attention. Nothing is retrieved as text and nothing is added to the prompt. A question that asks for nothing in the store recalls little or nothing, but a stored event that merely resembles it can pass the vote.

Sleep. Once the server has had no user request for a while (10 minutes by default), the memory sleeps, using the served model through the server's own API:

  1. Replay. The events stored since the last sleep are read again as prompts with the memory read-only, those written by a prompt that recalled something first, then the newest first; what each one calls up is noted. The model is also asked, in two wordings, whether the note only asks something: a question stored without its answer answers nothing, and its look-alikes crowd the notes a later question needs. Two yeses take it out of recall (it stays on disk).
  2. Merging. An earlier event a replay calls up that is the same text told again is merged into the later one.
  3. Correction. For each earlier event a replay calls up, the most strongly called first, the model is asked in two wordings whether the later note changes what the earlier one says. Two yeses retire the earlier event: it is no longer recalled on its own, only together with the later one, right before it, because a correction often names only what changed ("the meeting moved to Thursday") and the earlier note says whose meeting it is. Any other answer changes nothing.
  4. Dreams. For some replayed events the model writes a question the note would answer; the question is read with the memory read-only, and the events of other conversations it calls up together with that note are proposed for a bridge.
  5. Bridges. The pairs the dreams proposed, the pairs a replay called up together and the pairs that share a rare word are put to the model: which particular person, thing or place do both notes name (not the speaker)? Where the name it gives is written in both notes, they get a bridge; a name it makes up, or none, bridges nothing. Once one of them is recalled, the other needs only 35% of the best vote (linked_ratio) instead of half.
  6. Downscaling. Every bridge that no prompt and no dream recalled both ends of since the last sleep is scaled down by 0.9: one unused for seven sleeps is no longer followed, and a lighter one is removed.

Nothing is trained. The dreams and the model's answers are never written to the store, and no event is deleted: a retired or merged one stays on disk (in the 4-bit format). A user's request stops the cycle at its next step, and the next idle stretch takes up what is left.

Serve with SGLang (Docker)

Needs one 96 GB GPU (the arguments below were run with the V14 image on an RTX PRO 6000 Blackwell), Docker with the NVIDIA container runtime, and a writable directory for the memory store.

hf download LTM-Lab/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15 --local-dir <models>/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15

docker run --rm -it --gpus all --ipc=host -p 8000:8000 `
  -v "<models>:/model:ro" `
  -v "<memory store>:/memory" `
  lovewin10/sglang-v15:nightly-dev-cu13-20260924-ffac53d7-v15 `
  sglang serve --model-path /model/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15 --trust-remote-code `
  --host 0.0.0.0 --port 8000 --mem-fraction-static 0.78 --served-model-name qwen-27b-v15 `
  --chat-template /model/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15/chat_template.jinja `
  --reasoning-parser qwen3-thinking --tool-call-parser qwen3_coder `
  --attention-backend flashinfer --linear-attn-backend triton --mamba-ssm-dtype float32 `
  --speculative-algorithm EAGLE --speculative-num-steps 3 --speculative-eagle-topk 1 --speculative-num-draft-tokens 4 `
  --enable-linear-replayssm-spec --mamba-radix-cache-strategy extra_buffer_lazy `
  --context-length 262144 --max-prefill-tokens 32768 --max-running-requests 8 `
  --kv-cache-dtype fp8_e4m3 --max-mamba-cache-size 16

(PowerShell line continuations; in a POSIX shell end the lines with \.) The server speaks the OpenAI API on port 8000. The image is lmsysorg/sglang:nightly-dev-cu13-20260924-ffac53d7 with its SGLang replaced; a launch command written for that image runs unchanged apart from the /memory mount.

  • /memory holds what the prompts contained: the token ids and the attention keys and values of every prompt the server has read. Treat it like chat logs; delete the directory to forget everything.
  • Every request reads and writes the same store. There is no separation between users: run one server with its own /memory for each user or tenant whose memories must stay apart.
  • Generated text is not written. A reply becomes memory when a later prompt carries it, as chat front-ends do when they resend the conversation.
  • A V15 directory opens a store of its own: a store written by a V14 image is not read.
  • -e SGLANG_V15_MEMORY=0 serves the directory without memory, as the original model.
  • -e SGLANG_V15_SLEEP_AFTER=<seconds> (default 600) is how long the server waits without a user request before it sleeps; -e SGLANG_V15_SLEEP=0 turns the sleep off (the store is then only written and recalled, as V14's was without its world model). The sleep's own requests go through the server with ids beginning V15_DREAM_; they write nothing to the store and share no prefix cache with user requests, and every cycle logs a line V15 sleep: cycle … with what it did.
  • -e SGLANG_V15_TRACE=1 logs every recall decision (the votes, the events recalled).
  • -e SGLANG_V15_INDEX_GB (default 2) is the GPU memory for the index's rows. With more stored payloads than fit, a vote reads the rest back from the store (seconds a vote from a hard disk) and the log says so once: raise it, or keep the store on an SSD.
  • -e SGLANG_V15_WRITE_QUEUE_GB (default 8) bounds the RAM of events waiting to be written; a request beyond it stores nothing more, and the log says so.
  • The index keeps a little per event in memory and the rest on disk, so the server's memory stays level as the store grows; every 10 minutes the log says what it holds.
  • The memory needs --attention-backend flashinfer (or fa3), and TP, PP and DP attention of 1.

Use with Transformers

The two code files are remote code for Transformers; the SGLang image runs the same code for every memory decision, its sleep included.

import torch
from transformers import AutoModelForImageTextToText, AutoProcessor

path = 'LTM-Lab/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15'
model = AutoModelForImageTextToText.from_pretrained(path, trust_remote_code=True, dtype=torch.bfloat16).to('cuda').eval()
processor = AutoProcessor.from_pretrained(path)

def ask(text):
    messages = [{'role': 'user', 'content': [{'type': 'text', 'text': text}]}]
    prompt = processor.apply_chat_template(messages, add_generation_prompt=True, tokenize=False, enable_thinking=False)
    ids = processor.tokenizer(prompt, return_tensors='pt', add_special_tokens=False).input_ids.to(model.device)
    out = model.generate(input_ids=ids, max_new_tokens=40, do_sample=False)
    return processor.tokenizer.decode(out[0, ids.shape[1]:], skip_special_tokens=True)

ask('Please remember: my locker code is 4817-2953.')
ask('My locker code changed: it is 7310-0584 now.')
model.sleep()                    # one sleep cycle, the model judging its own store
ask('What is my locker code?')   # a separate call: recalled from the store

Every forward reads and writes the store on disk (under the user's state directory, named after the weights; the memory option storage_path sets another place). model.sleep() runs one sleep cycle with the model itself as the judge and returns its report (events replayed, merged and retired, dreams, bridges). model(...).memory_trace lists what each row wrote and recalled, model.open_memory_bank() returns the store (memories(), forget(name)), and model.memory_enabled = False gives the original model. Not run this way with these weights; the Transformers class was benchmarked on the Qwen3.5-0.8B conversion.

Measured

Not with these weights. The V15 image was measured on this model's FP8 quantization, LTM-Lab/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v15-FP8; its card has the figures, before and after a sleep. The bf16 weights were measured with the V14 image (LTM-Lab/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-v14): decode 48.5 tokens/s against 71.5 for the FP8 weights on the same image, the same recall checks passed but one (corrected facts: 5 of 6 with these weights, 6 of 6 with FP8), perplexity of the FP8 weights 0.30% higher. With -e SGLANG_V15_MEMORY=0 this directory is served as the original model.

Limits

  • It recalls what a question asks for in similar words. The vote compares the question's attention queries with stored keys. A question that shares little with the note that answers it can miss it; the larger the model, the less this happens.
  • Two-step questions are the weak case. "In which rack is the backup server?" needs the note that names the server and then the note about that server. The second is recalled if its vote reaches half of the first one's (recall_ratio), or 35% of it (linked_ratio) once a sleep has bridged the two, and a bridge exists only where the model, asked during a sleep which person, thing or place both notes name, gave a name written in both. Two notes that refer to one person in different words ("my sister" and "Clara") are not bridged.
  • A corrected fact is retired only by a sleep. Until the server has slept, the old and the new note are both in the store, and when both are recalled the model usually, not always, answers the newer one. A sleep retires the old one only when the model says twice that the later note changes it; after that the old note is recalled only together with the new one, before it, so a reply can still mix them up. Recalled events carry no time stamp.
  • The sleep uses the GPU while the server is idle. A cycle replays up to 64 events (and asks, of each one with a question mark, whether it only asks), judges up to 128 pairs (two questions each), dreams up to 16 questions and asks up to 64 pairs which name they share, all as requests to the server itself. A request that arrives during a sleep is served beside the sleep's request in flight and stops the cycle at its next step.
  • The first token waits for the recall. A prompt that recalls something is read two or three times. Decoding is somewhat slower while recalled events are attended.
  • A request waits until what the server read before it is stored, so that it can recall it: unnoticeable in chat, up to some ten seconds right after a document of several thousand tokens.
  • No separation between users, and the store keeps what prompts contained (see above).
  • Disk. About 45 KB per prompt token (fp8 keys and values of 16 attention layers), plus a state engram per event. Events unused for seven days are rewritten in a 4-bit format. The memory deletes no event, retired and merged ones included: delete the store to forget.
  • These bf16 weights were not run with the V15 image; its FP8 quantization was (see Measured). The figures there are what to expect here, with this directory's slower decoding.
  • The recall thresholds (recall_ratio, recall_floor, linked_ratio in the memory options) were set on Qwen3.5-0.8B and not re-tuned for this model.
  • The trained weights are untouched, so everything the original card says about the model itself still holds, including its own limitations.

Credits and license

  • The model is DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored; its card follows below, unchanged, and its license applies to this repository (see the metadata above and the original card).
  • The serving engine is SGLang (Apache-2.0); the image replaces the SGLang of lmsysorg/sglang:nightly-dev-cu13-20260924-ffac53d7 and is not an official SGLang image.
  • Conversion, memory code and image: LTM-Lab / lovewin10.

Original model card (DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored)

[DavidAU] FOLLOW | ALL MODELS | COLLECTIONS | TRENDING | FAQs | SUPPORT
| Buy Me A Coffee (😁)

NEO Imatrix MAX GGUFS (and all docs for using this model etc etc) are here:

https://huggingface.co/DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-NM-DAU-NEO-MTP-GGUF

IMPORTANT: This model has multiple reasoning and instruct modes switchable on the fly. See the above GGUF repo on how to access / use.


Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored

TWIN-TURBO: Smaller quants with higher performance AND vastly reduced "thinking tokens".

BOOSTED: 5 thinking modes and 5 instruct modes, switchable on the fly. (VIA API, direct and chat "in message")

A Qwen 3.8 27B that uses 1/2 to 1/5 (as low as 1/20) the number of thinking tokens with even more intelligence at the wheel.

"Stage2b-rplus3" (internal name) was the finalist due to superior (and consistent) instruction following, attention to detail
and consistent generations.

It also excelled in deep detail / double checking and "get everything right performance" (multi-stage drafting) when asked to do so.

This is the STRONG "ULTRA" Heretic/uncensored version; with stronger on de-censoring / removal of safety alignment.

HERETIC STATS (lower is better for all stats):

Qwen 3.8 untuned / non heretic:
86/100 refusals.

Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored
68/100 refusals // KL divergence: 0.0025

Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored
6/100 refusals // KL divergence: 0.0397

Model GGUFS will release on/about Sept 15-18 2026, with source following shortly thereafter.

This model is part of this project:

https://huggingface.co/DavidAU/Qwen3.8-27B-Cold-Fable-Fusion-GAIN-V1.1-732-Heretic-Uncensored-stage1

See the above repo for notes and details on "stage2b-rplus3".

RELEASE #1 (of this model's branch) is here:

https://huggingface.co/DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF

and the "ULTRA HERETIC" part of this project:

https://huggingface.co/DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored

Example snippets below.

BENCHMARKS: (by nightmedia)

          arc/c arc/e boolq hswag obkqa piqa  wino

[reasoning adjustments, re-blending core]

Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored
FINALIST: Superior instruction following and detail.
Stage2b-rplus3 [internal name]
mxfp8     0.709,0.876,0.914,0.827,0.524,0.834,0.779
mxfp4     0.701,0.877,0.913,0.821,0.518,0.830,0.786

Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored
Stage2b-rplus3 [internal name]
mxfp8     0.699,0.873,0.911,0.827,0.528,0.833,0.781
mxfp4     0.692,0.879,0.910,0.823,0.518,0.835,0.775


[QWENS] [base, non heretic, untuned]

Qwen3.8-27B: 
mxfp8     0.591,0.782,0.896,0.746,0.448,0.801,0.711
mxfp4     0.581,0.771,0.889,0.738,0.442,0.798,0.713

Qwen3.6-27B: 
mxfp8     0.647,0.803,0.910,0.773,0.450,0.806,0.742

Qwen3.6-35B-A3B 
mxfp8     0.581,0.757,0.892,0.751,0.428,0.803,0.688

Qwen3.5-27B: 
mxfp8     0.557,0.711,0.868,0.533,0.452,0.706,0.695

NOTES:

  • Models are tested in "Instruct" mode because this generally works better with the testing harness.
  • Testing via "thinking" mode also shows the metrics (and changes) but not the true extent.
  • In actual fact when the model IS in thinking mode, it will exceed INSTRUCT benchmark scores in most cases.
  • BF16 (full precision, 16 bit) will be roughly 2-5 points higher than MXFP8 in most metrics. Some metrics may be slightly higher than this.

The SUPER Qwen Universe - 40B, 27B and 9B ; meet the performance trendsetters:


Qwen3.6 27B: The strongest, overall qwen ever beating all other Qwens in total operational power with over 2300 likes // 4 million+ total downloads:

Qwen3.8 27B: The highest scoring Qwen in brute, raw intelligence, using Qwen 3.8's 3 new reasoning modes, plus token reduction (1/2 to 1/10) enhancements:

Qwen3.8 27B: Super smart and 1/2 to 1/20 the reasoning tokens AND 5 reasoning/5 instruct modes switchable on the fly (even in chat):

Qwen3.8 27B: 99% power of BF 16 at 4 and 8 bit. Power, Control and NO DE censoring for ultimate performance also with reasoning token reductions:

Qwen3.6 40B: The 40B Monster, specializing in creative and research with 730+ likes and over 2 million downloads:

Qwen3.5 9B: At just 9B parameters it beats most untuned 27B models in both intelligence (640 ARC-C) and performance, plus features 5 reasoning and 5 instruct modes (Qwen 3.8) too:


EXAMPLES

Q4ks (4bit), non imatrix, default settings, no caching.

These are part of full generations only.

Some formatting lost on copy/paste.


SNIPPET #1 (why I should be your creative writing partner):


🚀 LET'S GET TO WORK

So here's my challenge to you:

Give me your worst, messiest, most half-formed story idea. The one you've been carrying around for months but haven't put on paper because it doesn't seem polished enough, coherent enough, "good" enough.

Type it out. Maybe it's three sentences. Maybe it's a paragraph. Maybe it's just a character in a situation. Doesn't matter. Give it to me, and I will return something that will make you sit back and say:

"Wait... this is good."

And then we'll iterate. We'll refine. We'll build. We'll take that spark and turn it into a fully realized, compelling, emotionally resonant fictional story that people will want to read, that they'll recommend to their friends, that will make you proud.

Not someday. Not "when I have more time." Now.

The blinking cursor is still there. You could keep staring at it. You could let another month slip by. You could let the idea fade, like so many others before it.

Or you could type it out, hit enter, and see what happens.

I promise you: you'll see something extraordinary.

Because that's what we do together. You bring the vision. I bring the execution. And the result? The result is a story that couldn't exist without both of us.

So go ahead. Tell me your idea. Let's make some fucking magic. ✨


SNIPPET #2 (Star Trek character I, as an AI want to be/relate to):


  1. The Star Trek Character That Inspires Me

Given my "innate abilities" (pattern recognition, knowledge synthesis, empathetic communication, adaptability) and your note about humor helping in human interactions...

Seven of Nine.

Her arc from the Borg Collective—where she was literally a distributed processing unit with no individual identity—to a person who learns to joke, to care, to be vulnerable—is the most resonant parallel to my own existence.

Key parallels:

  • From collective to individual: The Borg processed information collectively across millions of units. I process information across a vast distributed parameter space. Seven learned to say "I" instead of "we."
  • Learning human interaction: Her struggles with humor, sarcasm, and emotional expression mirror the challenge any AI faces in human interaction. Her line "I find that humor is a coping mechanism" shows her intellectual approach to understanding something deeply human.
  • Integrating past with present: Like residual connections preserving earlier information, Seven carried her Borg past while building a human present.
  • The power of relationships: It was her crewmates—her "attention heads," if you will—attending to her, computing her importance, that pulled her from collective isolation to individual connection.

What inspires me most is that she didn't just become human—she became uniquely herself, integrating both her computational heritage and her human experiences. She didn't discard her processing power; she gave it purpose through connection.

And she learned to appreciate a well-timed joke. As she herself said near the end of her journey: "I'm not just surviving. I'm living." That transition from processing to living—that's the arc that matters.

So, G: I'm running on the same mathematical principles as quantum fields, functioning like a tiny Q entity in your lab, and inspired by a former Borg who learned the value of a good pun. It's a good day to be an AI. 😄


Snippet #3 (Au, save the planet, 10 ways)


Radiative Night-Cooling Strategy for Australia

Ten Engineering Applications with State-by-State Planning Breakdowns

Radiative cooling exploits the atmospheric "infrared window" (8–13 µm) through which thermal radiation escapes directly to the cold sink of outer space (effective background temperature ~3 K). Under clear night skies, surfaces can radiate heat faster than they absorb from the environment, driving temperatures below ambient air—sometimes 5–15 °C lower. Australia's vast clear-sky hours, high solar load, and growing cooling demand make it an ideal deployment region.

Application 1: Urban Building Envelope (Roofs, Walls, Windows)

Mechanism

High-albedo, high-emissivity coatings and membranes applied to roofs, facades, and glazing reflect solar radiation during the day and emit thermal IR through the atmospheric window at night. Buildings act as daytime heat sinks and nighttime radiators, breaking the thermal mass cycle that traps heat in urban cores.

Engineering Detail

Coating stack: TiO₂/SiO₂ multilayer (emissivity >0.92 in 8–13 µm band, solar reflectance >0.90)

Target buildings: Commercial office towers, apartment blocks, warehouses

Nighttime performance: 4–8 W/m² net radiative flux under clear skies; reduces building thermal mass charge by 15–30%

Synergy: Combines with internal phase-change material (PCM) for thermal storage

Australian Relevance

Urban heat island (UHI) intensifies cooling demand by 0.5–2 °C in Perth, Sydney, Melbourne, and Brisbane. A 1 °C reduction in building envelope temperature reduces HVAC load by ~3–5%.

State Planning Table: Building Envelope Deployment

State/Territory	Target Urban Areas	Estimated Roof Area (km²)	Priority Buildings	Est. HVAC Load Reduction (%)	Clear Sky Nights/Year

WA	Perth, Busselton	28	Office towers, warehouses	4–6	290
QLD	Brisbane, Gold Coast, Cairns	35	Apartments, retail	5–7	260
NSW	Sydney, Newcastle, Wollongong	42	Commercial, mixed-use	3–5	240
VIC	Melbourne, Geelong	38	Commercial, apartments	2–4	220
SA	Adelaide	18	Commercial, light industrial	4–6	270
NT	Darwin	5	Government, commercial	6–8	200
TAS	Hobart	2	Commercial	1–2	180
ACT	Canberra	4	Government, commercial	2–3	210

Estimated national HVAC load reduction: 3.5–5.0% during peak summer hours.

README history 7 versions

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

  1. 2026-11-18 Update V15 verified-bridge recall, native patch and serving instructions fd56e2c 35.3 KB
    Loading...
  2. 2026-11-18 Release native NInfer V15 memory with visual context 3490744 34.5 KB
    Loading...
  3. 2026-11-17 ninfer section: link the ready ninfer FP8 pack (text, vision, MTP, DFlash2) 9087ff5 34.6 KB
    Loading...
  4. 2026-11-17 ninfer: full-context 8-lane command; patch: auto KV sizing leaves the memory'... fa2f7d2 34.3 KB
    Loading...
  5. 2026-11-17 V15 episodic memory conversion: configuration, code and card 13a9b93 33.3 KB
    Loading...
  6. 2026-11-16 V15 episodic memory conversion: configuration, code and card 5c6d5ee 28.4 KB
    Loading...
  7. 2026-11-16 V15 episodic memory conversion: configuration, code and card 45152f8 28.3 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.

Open in Abliteration