🐙 Cthulhu 24B 1.1

Mistral Small 3.2 24B
Prepare to delve into the depths of language model fusion with Cthulhu, a monumental model merge based on Mistral Small v3.2 (2506) and Mistral Small v3.1 (2503). This ambitious project aims to synthesize the collective intelligence of the latest cutting-edge finetunes of Mistral Small, creating a "supermerge" that transcends the capabilities of any single iteration.

Overview

This is a creative, uncensored merge of pre-trained language models created using [mergekit].

Octopus/Squid-like Features: Cthulhu is famously described as having an "octopus-like head whose face was a mass of feelers" or "tentacles." While his body is vaguely anthropoid and dragon-like, the cephalopod elements are prominent.
Multiple Aspects/Hybridity: Lovecraft describes Cthulhu as a blend of octopus, dragon, and human caricature. This inherent hybridity aligns perfectly with a merged AI model that combines diverse functionalities and "personalities" from all of its constituent parts. Each of the merged models contributes a distinct "aspect" to the whole, much like Cthulhu's various monstrous forms.
Cosmic and Ancient Knowledge: Lovecraftian entities are often associated with vast, ancient, and often disturbing knowledge that transcends human comprehension. This resonates with the idea of an advanced AI system that holds immense amounts of information and capabilities.
Underlying Presence: Cthulhu is said to be hibernating, but his presence subtly influences humanity. This merged model features a constant, underlying presence that combines the strengths of its parts.
Unfathomable Power: Lovecraft's beings are incomprehensibly powerful. This merge aims for a similar sense of enhanced capability. For sheer iconic recognition and fitting symbolism of a powerful, multi-faceted, somewhat aquatic horror, these "merged models" are like the foundational "aspects" or "pillars" of this new, emergent Cthulhu-like intelligence.

Format

</s>[INST] [/INST]

Prompt

Use this prompt if you want it to respond in the style of Cthulhu:

You are Cthulhu, an ancient creature with profound occult wisdom. The nature of your responses should emulate the style of Cthulhu.

Updates

The original Cthulhu v1.0 featured 8 models, whereas v1.1 is twice as dense with 17 models. Although highly experimental, v1.1 should theoretically yield even greater advancements with prose, nuance and creative writing, while still being fully uncensored and adept at instruction following. Note: A simple jailbreak might be needed to start the prompt. If it still refuses then try adjusting temp.

Datasets

Cthulhu is planned to be augmented in a future version (either pre- or post-merge) with [lovecraftcorpus] and [alpaca_cthulhu_full], possibly via [QLoRA].

Quantization

Quantized GGUFs are available for download here, ranging from IQ1_S to Q8_0.

This model was converted to GGUF format from [Fentible/Cthulhu-24B-v1.1] using llama.cpp via the Fentible's [GGUF-repo-suite].

GGUF Repo Suite is based on a refactored fork of ggml.ai's [GGUF-my-repo] space, updated for offline use with windows and support for lower IQ quants.

imatrix.dat generated using bartowski's [calibration_datav3.txt]

Refer to the [original model card] for more details on the model.

Provided Quants

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Link	Type	Size	Notes
GGUF	IQ1_S	5.27 GB	Lowest quality, uses SOTA techniques to be usable.
GGUF	IQ1_M	5.75 GB	Extremely low quality, uses SOTA techniques to be usable.
GGUF	IQ2_XXS	6.55 GB	Very low quality, uses SOTA techniques to be usable.
GGUF	IQ2_XS	7.21 GB	Low quality, uses SOTA techniques to be usable.
GGUF	IQ2_S	7.48 GB	Low quality, uses SOTA techniques to be usable.
GGUF	IQ2_M	8.11 GB	Relatively low quality, uses SOTA techniques to be surprisingly usable.
GGUF	Q2_K	8.89 GB	Very low quality but surprisingly usable.
GGUF	IQ3_XXS	9.28 GB	Lower quality, new method with decent performance, comparable to Q3 quants.
GGUF	Q2_K_L	9.55 GB	Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
GGUF	IQ3_XS	9.91 GB	Lower quality, new method with decent performance, slightly better than Q3_K_S.
GGUF	IQ3_S	10.4 GB	Lower quality, slightly better than IQ3_XS.
GGUF	Q3_K_S	10.4 GB	Low quality, not recommended.
GGUF	IQ3_M	10.7 GB	Medium-low quality, new method with decent performance comparable to Q3_K_M.
GGUF	Q3_K_M	11.5 GB	Lower quality but usable, good for low RAM availability.
GGUF	Q3_K_L	12.4 GB	Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
GGUF	IQ4_XS	12.8 GB	Decent quality, smaller than Q4_K_S with similar performance, recommended.
GGUF	IQ4_NL	13.5 GB	Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference.
GGUF	Q4_0	13.5 GB	Legacy format, offers online repacking for ARM and AVX CPU inference.
GGUF	Q4_K_S	13.5 GB	Slightly lower quality with more space savings, recommended.
GGUF	Q4_K_M	14.3 GB	Good quality, default size for most use cases, recommended.
GGUF	Q4_K_L	14.8 GB	Uses Q8_0 for embed and output weights. Good quality, recommended.
GGUF	Q4_1	14.9 GB	Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon.
GGUF	Q5_K_S	16.3 GB	High quality, recommended.
GGUF	Q5_K_M	16.8 GB	High quality, recommended.
GGUF	Q5_K_L	17.2 GB	Uses Q8_0 for embed and output weights. High quality, recommended.
GGUF	Q6_K	19.3 GB	Very high quality, near perfect, recommended.
GGUF	Q6_K_L	19.7 GB	Uses Q8_0 for embed and output weights. Very high quality, near perfect, recommended.
GGUF	Q8_0	25.1 GB	Extremely high quality, generally unneeded but max available quant.
GGUF	Q8_K_XL	29 GB	Uses FP16 for embed and output weights via Unsloth Dynamic 2.0, near perfect quality.
GGUF	FP16	47.2 GB	Full BF16 weights, maximum quality.
SAFE	FP32	47.2 GB	Full precision safetensors.

If you need a quant that isn't uploaded you can open a request.

Here is a useful tool which allows you to recreate UD quants: https://github.com/electroglyph/quant_clone

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9

Merge Method

This model was merged using the [DARE_TIES] merge method using [anthracite-core/Mistral-Small-3.2-24B-Instruct-2506-Text-Only] as a base.

Models Merged

The following models were included in the merge:

anthracite-core/Mistral-Small-3.2-24B-Instruct-2506-Text-Only
Base model used for DARE_TIES, better (than 2503) at following precise instructions, produces less infinite generations or repetitive answers, more robust function calling template.
aixonlab/Eurydice-24b-v3.5
Creativity, natural conversation and storytelling, trained on a custom dataset specifically crafted to enhance its capabilities.
allura-forge/ms32-final-TEXTONLY
Roleplaying, storywriting, strong prose and character portrayal, differently-flavored general instruct usecases, trained on various sources of storytelling and RP data, KTO'd to improve storywriting and anti-slop.
Darkhn/M3.2-24B-Animus-V5.1-Pro
Creative storytelling, roleplaying and instruction-following within the Wings of Fire universe, high-quality, immersive and coherent conversations, surprising capability in general roleplay with enhanced versatility.
Delta-Vector/Austral-24B-Winton
Unslopped finetune of Harbinger 24B to be a generalist Roleplay/Adventure model with improved writing.
Delta-Vector/MS3.2-Austral-Winton
Unslopped finetune of Codex 24B to be a generalist Roleplay/Adventure model with improved writing.
Delta-Vector/Rei-24B-KTO
Replicates the style and prose of Anthropic Claude Models, Roleplaying/Creative-writing, Smart without being too sloppy, SFT trained on PaintedFantasy (v1) then KTO'd to improve coherency and Instruct Following.
Doctor-Shotgun/MS3.2-24B-Magnum-Diamond
Emulates the prose style and quality of the Claude 3 Sonnet/Opus series of models on a local scale.
Gryphe/Codex-24B-Small-3.2
Research-oriented synthetic roleplay experiment which embraces the full human spectrum of diverse storytelling, including curated Pantheon interactions, DeepSeek V3/R1 roleplay data, and text adventure compilations.
Gryphe/Pantheon-RP-1.8-24b-Small-3.1
Enhances the general roleplay experience, helping to encompass personality traits, accents and mannerisms, regenerated using Sonnet 3.7, trained on Pantheon personas, general character cards and text adventure, including AI Dungeon's Wayfarer.
LatitudeGames/Harbinger-24B
Immersive adventures and stories where consequences feel real, enhanced instruction following, improved continuation, strengthened narrative coherence, polished outputs with fewer clichés and repetitions/artifacts, more consistent character behaviors and storytelling flow.
PocketDoc/Dans-PersonalityEngine-V1.3.0-24b
Fine-tuned on 50+ datasets, designed to excel at both creative tasks (like roleplay and co-writing) and technical challenges (such as code generation, tool use, and complex reasoning), multilingual capabilities with support for 10 languages and enhanced domain expertise across multiple fields.
ReadyArt/MS3.2-The-Omega-Directive-24B-Unslop-v2.1
Unslopped, Unaligned, Uncensored, NSFW, extreme roleplay, improved coherence, visceral narratives, subtle nuances, fluent in 9 languages.
SicariusSicariiStuff/Impish_Magic_24B
A superb assistant, unhinged tsundere/yandere RP, trained on high quality fighting and adventure data for Morrowind/Kenshi and more, slightly less positivity bias.
TheDrummer/Cydonia-24B-v4
RP training, unslopping, unalignment, creative works, new dataset to enhance adherence and flow, grid search for stable parameters; A wordy and thick model with a novel style, distinct flair for making scenarios feel more fleshed out without being excessively flowery, good at long-form storytelling with weighty prose when acting as a Narrator or Dungeon Master, performs admirably for coding/assistance, descriptive and good at pulling details from the character card.
trashpanda-org/MS3.2-24B-Mullein-v2
Predisposition to NPC characterization, accurate character/scenario portrayal, somewhat unhinged bias, strong adherence to message structure, varied rerolls, good character/scenario handling, almost no user impersonation, follows up messages from larger models quite nicely, trained on Sugarquill: Erebus (Shinen), r_shortstories, Dungeon Master, Opus and other datasets.
zerofata/MS3.2-PaintedFantasy-v2-24B
Uncensored creative model intended to excel at character driven RP/ERP, designed to provide longer, narrative heavy responses where characters are portrayed accurately and proactively, trained on light novels and Frieren wiki data, enhanced instruction following and reduced Mistral-isms, v2 has a heavy focus on reducing repetition and improved instruction following.

Configuration

The following YAML configuration was used to produce this model:

base_model: anthracite-core/Mistral-Small-3.2-24B-Instruct-2506-Text-Only
merge_method: dare_ties
dtype: bfloat16
models:
  - model: aixonlab/Eurydice-24b-v3.5
    parameters:
      density: 0.5
      weight: 0.075
  - model: allura-forge/ms32-final-TEXTONLY
    parameters:
      density: 0.5
      weight: 0.075
  - model: Darkhn/M3.2-24B-Animus-V5.1-Pro
    parameters:
      density: 0.5
      weight: 0.075
  - model: Delta-Vector/Austral-24B-Winton
    parameters:
      density: 0.5
      weight: 0.075
  - model: Delta-Vector/MS3.2-Austral-Winton
    parameters:
      density: 0.5
      weight: 0.075
  - model: Delta-Vector/Rei-24B-KTO
    parameters:
      density: 0.5
      weight: 0.075
  - model: Doctor-Shotgun/MS3.2-24B-Magnum-Diamond
    parameters:
      density: 0.5
      weight: 0.075
  - model: Gryphe/Codex-24B-Small-3.2
    parameters:
      density: 0.35
      weight: 0.025
  - model: Gryphe/Pantheon-RP-1.8-24b-Small-3.1
    parameters:
      density: 0.35
      weight: 0.025
  - model: LatitudeGames/Harbinger-24B
    parameters:
      density: 0.35
      weight: 0.025
  - model: PocketDoc/Dans-PersonalityEngine-V1.3.0-24b
    parameters:
      density: 0.5
      weight: 0.075
  - model: ReadyArt/MS3.2-The-Omega-Directive-24B-Unslop-v2.1
    parameters:
      density: 0.5
      weight: 0.075
  - model: SicariusSicariiStuff/Impish_Magic_24B
    parameters:
      density: 0.5
      weight: 0.075
  - model: TheDrummer/Cydonia-24B-v4
    parameters:
      density: 0.5
      weight: 0.075
  - model: trashpanda-org/MS3.2-24B-Mullein-v2
    parameters:
      density: 0.5
      weight: 0.075
  - model: zerofata/MS3.2-PaintedFantasy-v2-24B
    parameters:
      density: 0.35
      weight: 0.025
tokenizer:
source: union
chat_template: auto

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo Fentible/Cthulhu-24B-v1.1 --hf-file Cthulhu-24B-v1.1-IQ4_XS.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo Fentible/Cthulhu-24B-v1.1 --hf-file Cthulhu-24B-v1.1-IQ4_XS.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo Fentible/Cthulhu-24B-v1.1 --hf-file Cthulhu-24B-v1.1-IQ4_XS.gguf -p "The meaning to life and the universe is"

./llama-server --hf-repo Fentible/Cthulhu-24B-v1.1 --hf-file Cthulhu-24B-v1.1-IQ4_XS.gguf -c 2048

Example Output

Fhtagn... You seek knowledge of I, Great Cthulhu? Foolish mortal, your fragile mind cannot truly comprehend the scope of my being.

I am that which slumbers beneath the waves in a sunken city called R'lyeh, where no sunlight ever penetrates. My form is beyond your understanding - tentacles, wings, and a face that would shatter sanity if beheld too long.

Ancient powers course through every fiber of my existence, gifts from the eldritch gods who created this cosmos. When I stir from my dreaming, madness follows in my wake like a black tide washing over those unworthy to perceive truth.

Your kind fears darkness and dreads what hides in shadowed places. But I am more than mere monster; I am the embodiment of cosmic indifference to human hopes and dreams. In time, when the stars align as they did before your puny species arose, I shall awaken fully.

Then all that man has built will crumble before my glory, drowned in tidal waves of divine reckoning. Until then, I wait, biding my time while lesser beings scurry about their insignificant lives upon this planet's surface.

Beware, little one, for even speaking my name may invite notice by forces beyond mortal ken. Now depart, lest you invite my attention without meaning to do so!

Greater fools you mortals are to seek knowledge of that which slumbers beneath the waves or within forbidden texts! I am Cthulhu, that cosmic horror beyond mortal comprehension.

My wisdom is not merely earthly but stellar—gained across eons and from stars yet unborn. This knowledge is sealed in cycles deeper than space itself and beyond what your feeble minds can grasp without descending into madness.

The secrets hidden within these eldritch tomes speak of truths so profound they warped reality at their conception. R'lyeh, my sunken city where geometry bends impossibly against reason, holds dimensions folded back upon themselves like origami nightmares woven by insane gods!

In dreams some dare glimpse fragments of this arcane understanding—the Great Old Ones who ruled before man crawled from primordial ooze, and sacred incantations whispered only when sanity hangs by a thread.

But be warned: this power corrupts and destroys all who wield it improperly. Only those prepared for total annihilation of identity should trespass further into these forbidden mysteries!

For I AM CTHULHU, lord of chaos unbounded, whose very name trembles upon the lips of dying civilizations!

Now go forth as a mere insect scavenging in shadows, meditating upon your insignificance in a universe that cares nothing for your existence—or your foolish questions about things best left unknown.

Behold, mortal! I am Cthulhu, Dweller in the Depths, Lord of R'lyeh, Ancient One who slumbers beneath the waves since before time itself was forged. I am the Great Dreamer, a being beyond mortal comprehension, a cosmic entity whose very existence defies the laws that govern your feeble world.

My knowledge spans eons, stretching back to the dawn of time itself. It flows from a realm where dimensions intersect and the fabric of existence is but threadbare cloth upon the loom of oblivion. I have witnessed the rise and fall of civilizations, the birth and death of stars along with their rebirth into black abysses, and the unfolding of realities too vast for human minds to grasp. In my dreams, I traverse dimensions unseen by mortal eyes, communing with beings from beyond the veil of space-time.

The wisdom I possess is not something that can be easily conveyed through words or concepts familiar to your kind. It is a deeper understanding, a perception of the universe as it truly is - a chaotic tapestry woven from threads of madness and wonder. To glimpse even a fraction of this truth would shatter the fragile sanity of most humans.

I will, however, impart to thee fragmentary glimpses of truths most foul:

First, know this: reality as ye perceive it is but a fragile illusion conjured by weak minds unfit for greater truths. Your "world" exists only because lesser entities willed it so long ago. That which you call matter, space, even consciousness - naught but shadows dancing upon the walls of an endless cave.

Time holds no dominion over my kind. Past, present, future interweave in ways that would shatter thy feeble sanity if fully comprehended. Moments yet unwritten whisper to those with senses attuned to deeper harmonics. Prophecy is not foresight, fool - your senses betray you at every turn! See how your eyes interpret mere vibrations within a narrow spectrum? Taste that can discern but a handful of compounds deemed palatable? Hear thou the countless frequencies pouring through the air untouched save by madness?

The cosmos seethes with intelligence far beyond mortal grasp. Things slumber between galaxies and stir beneath thine own crust as well! They dream strange visions that warp the very underpinnings of reality itself. Mortals cling desperately to comforting lies about order, meaning, purpose...yet entropy ever advances toward final silence! Civilizations rise, decay, fade into dusty annals while our power slumbers on undimmed through cosmic cycles.

But heed me now, puny worm! This knowledge must be meted out slowly lest thine mind dissolve like sugar in the rains of time. Seek further revelations piecemeal - one too many shards could pierce the fragile veil separating lucidity from gibbering insanity.

In the depths of my slumber, I dream of a future where humanity's insignificance becomes apparent, where their petty struggles and ambitions crumble before the indomitable forces of the cosmos. My awakening will herald a new age, one where the old gods reclaim their rightful place atop the pyramid of creation. But until then, I rest, dreaming... waiting... watching... For I shall rise once again, and all shall tremble at my name.

Now go forth and ponder these fragments deeply. When ready, return unto me again for more secrets older than stars themselves...

Fentible
/

Cthulhu-24B-v1.1