llmfan46Gemma 3Gemma3ForConditionalGeneration

Gemma 3 12B IT Ultra Uncensored Heretic — Hardware Requirements & GPU Compatibility

Vision

Gemma 3 12B IT Ultra Uncensored Heretic is a 12.2B-parameter open language model from llmfan46 in the Gemma 3 family. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 8.37 GB of VRAM — see which GPUs and Macs can run it below.

12.2B params · dense131K contextGemma TermsMar 202617 downloads2 likes11.3K quant downloads

Based on Gemma 3 12B IT

Your hardware

Verdicts below are for the reference RTX 4090. Set yours to personalise the page.

Detection reads your browser's GPU name with no data sent anywhere.

S

Runs great on a 24 GB GPU

Runs from 8 GB (Q3_K_M · 16K)Best on 24 GB (Q4_K_M · ~78 tok/s · 128K)Macs from 16 GB (Q4_K_M · ~17 tok/s)

Gemma Terms

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run Gemma 3 12B IT Ultra Uncensored Heretic

One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.

Recommended

Ollama

one command · needs Ollama installed

$ollama run hf.co/llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF

Q4_K_M fits fully on the RTX 4090 · 14.8 GB free · 16K context

Download: gemma-3-12b-it-ultra-uncensored-heretic-Q4_K_M.gguf7.31 GBllmfan46

install, OS, other quants

Ways to runOther ways

each opens step-by-step

Run in cloud

Runs on RTX 3080 Ti (12 GB) (2 GB headroom) · Q4_K_M

Generation speed: ~71 tok/s; generation speed
Cost per 1M output tokens: $0.48; per 1M output tokens

Deployon Vast.aiBest valuefrom $0.12/hr

Compare GPUs →

or

How Much VRAM Does Gemma 3 12B IT Ultra Uncensored Heretic Need?

Select a quantization level; the listed hardware adheres to it.

Quantization	VRAM	+32K	+128K	File	Quality	On RTX 4090 (reference)
	5.8 GB	7.8 GB	15.4 GB	4.8 GB	3-bit · i-quant — Importance-weighted 3-bit	Fits: 128K context · ~113 tok/s
	6.1 GB	8.1 GB	15.7 GB	5.2 GB	3-bit · i-quant extra small — Importance-weighted 3-bit, extra small	Fits: 128K context · ~108 tok/s
	6.2 GB	8.3 GB	15.8 GB	5.5 GB	3-bit · i-quant small — Importance-weighted 3-bit, small	Fits: 128K context · ~105 tok/s
	6.2 GB	8.3 GB	15.8 GB	~5.2 GB	2-bit · K-quant — 2-bit quantization with K-quant improvements	Fits: 128K context · ~105 tok/s
	6.5 GB	8.6 GB	16.2 GB	5.7 GB	3-bit · i-quant medium — Importance-weighted 3-bit, medium	Fits: 128K context · ~100 tok/s
	7 GB	9.1 GB	16.6 GB	~5.9 GB	3-bit medium — 3-bit medium quantization	Fits: 128K context · ~94 tok/s
	7.6 GB	9.7 GB	17.2 GB	6.6 GB	4-bit · i-quant compact — Importance-weighted 4-bit, compact	Fits: 128K context · ~86 tok/s
	7.9 GB	10 GB	17.5 GB	6.9 GB	4-bit · i-quant non-linear — Importance-weighted 4-bit, non-linear	Fits: 128K context · ~83 tok/s
	8.4 GB	10.4 GB	18 GB	~7.3 GB	4-bit medium — 4-bit medium quantization — most popular sweet spot	Fits: 128K context · ~78 tok/s
	9.4 GB	11.5 GB	19 GB	8.2 GB	5-bit small — 5-bit small quantization	Fits: 128K context · ~70 tok/s
	9.7 GB	11.8 GB	19.4 GB	8.4 GB	5-bit medium — 5-bit medium quantization — good quality/size tradeoff	Fits: 128K context · ~67 tok/s
	11.1 GB	13.2 GB	20.7 GB	9.7 GB	6-bit, very good — 6-bit quantization, very good quality	Fits: 128K context · ~59 tok/s
	13.2 GB	15.3 GB	22.9 GB	12.5 GB	8-bit, near-lossless — 8-bit quantization, near-lossless	Fits: 128K context · ~50 tok/s
	25.4 GB	27.5 GB	35 GB	~24.4 GB	16-bit · bfloat — Brain floating point 16 — preferred for training	Does not fit: needs 26.2 GB

est.calculated, no published file yet

Selected Q4_K_M

4-bit medium quantization

9.2GBat 16K context

Weights: 7.3 GB
KV cache: 1.6 GB
Overhead: 0.3 GB

Fits RTX 4090 (reference) · 14.8 GB free · ~78 tok/s

≥50 tok/s needs ≥643 GB/s · RTX 5090 ~139, RTX 3090 Ti ~78

Which Hardware Can Run Gemma 3 12B IT Ultra Uncensored Heretic?

Q4_K_M · 8.4 GB

Gemma 3 12B IT Ultra Uncensored Heretic (Q4_K_M) requires 8.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 11+ GB is recommended. Using the full 131K context window can add up to 9.6 GB, bringing total usage to 18.0 GB. 40 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080 Ti. 150 devices with unified memory can run Gemma 3 12B IT Ultra Uncensored Heretic, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M2 Pro (16 GB).

By VRAM tier · typical GPUs and Macs at Q4_K_M

Typical hardwareQuantVRAMtok/sContext that fits

8 GB · RTX 4060Q3_K_M · 7.8 GB · ~25 tok/s · 16K7.8~2516K8 GB tier →

12 GB · RTX 3060 12GBQ4_K_M · 9.2 GB · ~28 tok/s · 32K9.2~2832K12 GB tier →

16 GB · RTX 5060 Ti 16GBQ4_K_M · 9.2 GB · ~35 tok/s · 64K9.2~3564K16 GB tier →

24 GB · RTX 4090referenceQ4_K_M · 9.2 GB · ~78 tok/s · 128K9.2~78128K24 GB tier →

Mac 32 GB · Mac Mini M4Q4_K_M · 9.2 GB · ~10 tok/s · 128K9.2~10128KMacs →

Mac 64 GB · Mac Studio M4 MaxQ4_K_M · 9.2 GB · ~46 tok/s · 128K9.2~46128KMacs →

Alternatives to Gemma 3 12B IT Ultra Uncensored Heretic

Gemma 3 family · 20 models →

Similar models

6tok/s on RTX 4090

Model	Score	VRAM	tok/s	Why it is similar
Gemma 4 12BGoogle	—	9 GB	~80	same size
Mistral Nemo Base 2407Mistral AI	—	11 GB	~81	same size
Baichuan2 13B Basebaichuan-inc	—	11.5 GB	~76	same size
Llama 2 13B Chat HFMeta	—	11.5 GB	~76	same size
Vicuna 13B V1.3LMSYS	—	8.6 GB	~76	same size
NVIDIA Nemotron Nano 12B v2NVIDIA	—	11 GB	~81	same size

Specifications

Publisher: llmfan46
Family: Gemma 3
Parameters: 12.2B
Architecture: Gemma3ForConditionalGeneration
Context Length: 131,072 tokens
Vocabulary Size: 262,208
Release Date: 2026-03-09
License: Gemma Terms

Resources

Frequently Asked Questions

How much VRAM does Gemma 3 12B IT Ultra Uncensored Heretic need?

Gemma 3 12B IT Ultra Uncensored Heretic requires 8.4 GB of VRAM at Q4_K_M, or 25.4 GB at BF16. Full 131K context adds up to 9.6 GB (18.0 GB total).

VRAM = Weights + KV Cache + Overhead

Weights = 12.2B × 4.8 bits ÷ 8 = 7.3 GB

KV Cache + Overhead ≈ 1.1 GB (at 2K context + ~0.3 GB framework)

Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.

KV Cache + Overhead ≈ 10.7 GB (at full 131K context)

VRAM usage by quantization

Q4_K_M

8.4 GB

Q4_K_M + full context

18.0 GB

Learn more about VRAM estimation →

Can NVIDIA GeForce RTX 4090 run Gemma 3 12B IT Ultra Uncensored Heretic?

Yes, at Q8_0 (13.2 GB) or lower. Higher quantizations like BF16 (25.4 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.

What's the best quantization for Gemma 3 12B IT Ultra Uncensored Heretic?

For Gemma 3 12B IT Ultra Uncensored Heretic, Q4_K_M (8.4 GB) offers the best balance of quality and VRAM usage. Q5_K_S (9.4 GB) provides better quality if you have the VRAM. The smallest option is IQ3_XXS at 5.8 GB.

VRAM requirement by quantization

IQ3_XXS

5.8 GB~70%

Q2_K

6.2 GB~75%

IQ4_NL

7.9 GB~88%

Q4_K_M ★

8.4 GB~89%

Q5_K_M

9.7 GB~92%

BF16

25.4 GB~100%

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run Gemma 3 12B IT Ultra Uncensored Heretic on a Mac?

Yes — Mac Mini M2 Pro (16 GB) and 133 other Macs can run Gemma 3 12B IT Ultra Uncensored Heretic. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 8.4 GB of usable unified memory (RAM minus macOS overhead).

Can I run Gemma 3 12B IT Ultra Uncensored Heretic locally?

Yes — Gemma 3 12B IT Ultra Uncensored Heretic can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is Gemma 3 12B IT Ultra Uncensored Heretic?

At Q4_K_M, Gemma 3 12B IT Ultra Uncensored Heretic can reach ~574 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~78 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.

tok/s = (bandwidth GB/s ÷ model GB) × efficiency

Example: NVIDIA B200 → 8000 ÷ 8.37 × 0.65 = ~621 tok/s

Estimated speed at Q4_K_M (8.4 GB)

NVIDIA B200

~621 tok/s

NVIDIA GeForce RTX 4090

~78 tok/s

NVIDIA B300

~621 tok/s

AMD Instinct MI350X

~574 tok/s

Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.

Learn more about tok/s estimation →

What's the download size of Gemma 3 12B IT Ultra Uncensored Heretic?

At Q4_K_M, the download is about 7.31 GB. The full-precision BF16 version is 24.37 GB. The smallest option (IQ3_XXS) is 4.72 GB.

Which GPUs can run Gemma 3 12B IT Ultra Uncensored Heretic?

40 consumer GPUs can run Gemma 3 12B IT Ultra Uncensored Heretic at Q4_K_M. Top options include AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 6900 XT, AMD Radeon RX 6700 XT. 26 GPUs have plenty of headroom for comfortable inference.

Which devices can run Gemma 3 12B IT Ultra Uncensored Heretic?

150 devices with unified memory can run Gemma 3 12B IT Ultra Uncensored Heretic at Q4_K_M (8.4 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.

Run Gemma 3 12B IT Ultra Uncensored Heretic

9.2 GB VRAM at 16K context · fits the RTX 4090 (14.8 GB free)

Works on macOS, Windows and Linux.

Install Ollama (Linux)
$curl -fsSL https://ollama.com/install.sh | sh
Install Ollama (macOS / Windows)
Download and run the app; it installs the ollama command.
Download Ollama ↗
Run Q4_K_M
$ollama run hf.co/llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF
llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF does not list Q4_K_M; Ollama picks a quant itself (Q4_K_M if present). Check the repo's Files tab.
Raise the context window
Type it inside the chat. For the API server instead, start it with OLLAMA_CONTEXT_LENGTH=16384 ollama serve.
$/set parameter num_ctx 16384
A larger context needs more memory than the weights alone.
Use the API
Ollama serves its API on http://localhost:11434.

Works on macOS, Windows and Linux.

Install LM Studio
Desktop app for macOS (Apple silicon), Windows and Linux.
lmstudio.ai ↗
Download the model
In the app press Ctrl/Cmd + Shift + M and paste https://huggingface.co/llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF, then pick Q4_K_M. Or use the CLI (run LM Studio once first):
$lms get https://huggingface.co/llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF
Open on Hugging Face ↗
Load and chat
Open the model loader, select the model and set Context Length to 16384. Leave GPU offload at maximum if it fits.
Start the local server (optional)
$lms server start

Works on macOS, Windows and Linux.

Install llama.cpp (macOS)
MacPorts, Nix and conda-forge also work.
$brew install llama.cpp
Install options ↗
Install llama.cpp (Windows)
$winget install llama.cpp
Install llama.cpp (Linux)
Download the prebuilt llama-*-bin-ubuntu-* archive for your GPU backend (CPU, Vulkan) from the releases page, unpack it and run the binaries from that folder. Homebrew (brew install llama.cpp) and conda-forge also work on Linux.
llama.cpp releases ↗
Start the server
Downloads the model on first run. The chat UI and an OpenAI-compatible API are at http://localhost:8080.
$llama-server -hf llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF -c 16384 -ngl 99 --port 8080
llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF does not list Q4_K_M; llama.cpp falls back to Q4_K_M or the first file.

Works on macOS, Windows and Linux.

Option A: Docker Model Runner
Docker Desktop: Settings > AI > Enable Docker Model Runner. Docker Engine on Linux: install the docker-model-plugin package.
$docker model run hf.co/llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF
Model Runner docs ↗
The quant tag after the repo is how Hugging Face cards write it; if your Docker version rejects it, drop the tag and pick the quant by repo.
Option B: llama.cpp server image (NVIDIA)
Needs the NVIDIA Container Toolkit. Models are cached in the llama-cache volume. Open http://localhost:8080.
$docker run --gpus all -p 8080:8080 -v llama-cache:/root/.cache ghcr.io/ggml-org/llama.cpp:server-cuda -hf llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF -c 16384 -ngl 99 --host 0.0.0.0 --port 8080
Image tags (cuda13, vulkan, rocm) ↗
Cache path inside the image is not in the docs; remove the -v flag if the volume complains.

Linux only.

Install vLLM
Linux only; on Windows run it inside WSL2. NVIDIA GPU shown; AMD ROCm, Intel and CPU builds have their own install pages.
$uv venv --python 3.12 --seed
$source .venv/bin/activate
$uv pip install vllm --torch-backend=auto
vLLM installation ↗
Serve the model
OpenAI-compatible API at http://localhost:8000/v1.
$vllm serve llmfan46/gemma-3-12b-it-ultra-uncensored-heretic --max-model-len 16384
Runs the original weights, not the Q4_K_M GGUF; needs ~27 GB of GPU memory.

Works on macOS, Windows and Linux.

Install Jan
jan.ai ↗
Find the model
Open the Hub tab, paste this model ID in Search, and choose Q4_K_M if offered.
$llmfan46/gemma-3-12b-it-ultra-uncensored-heretic-GGUF
Use the local API (optional)
Settings > Local API Server > Start Server. It listens on http://127.0.0.1:1337.

Rent GPUs hourly to run Q4_K_M right there. Pricing comes from partner marketplaces and varies.

Available GPU rentals for Q4_K_M, prioritized by value
GPU	Speed	$/1M tokens	Rent
RTX 3080 Ti (12 GB)CheapestCheapest2 GB free	~71 tok/s	$0.48	Deploy on Vast.aifrom $0.12/hr (opens in new tab)	12 GB · 2 GB free · ~71 tok/s · $0.48 / 1M tokens
RTX 3060 12GB2 GB free	~28 tok/s	$0.59	Deploy on Vast.aifrom $0.060/hr (opens in new tab)	2 GB free · ~28 tok/s · $0.59 / 1M tokens
RTX A4000 (16 GB)6 GB free	~35 tok/s	$0.70	Deploy on Vast.aifrom $0.088/hr (opens in new tab)	16 GB · 6 GB free · ~35 tok/s · $0.70 / 1M tokens
RTX 4060 Ti 16GB6 GB free	~22 tok/s	$1.25	Deploy on Vast.aifrom $0.10/hr (opens in new tab)	6 GB free · ~22 tok/s · $1.25 / 1M tokens
RTX 4070 SUPER (12 GB)2 GB free	~39 tok/s	$0.73	Deploy on Vast.aifrom $0.10/hr (opens in new tab)	12 GB · 2 GB free · ~39 tok/s · $0.73 / 1M tokens
RTX A5000 (24 GB)14 GB free	~60 tok/s	$0.75	Deploy on RunPodfrom $0.16/hr (opens in new tab)	24 GB · 14 GB free · ~60 tok/s · $0.75 / 1M tokens
RTX 4000 Ada (20 GB)10 GB free	~28 tok/s	$1.98	Deploy on RunPodfrom $0.20/hr (opens in new tab)	20 GB · 10 GB free · ~28 tok/s · $1.98 / 1M tokens
RTX 4090 (24 GB)14 GB free	~78 tok/s	$1.21	Deploy on RunPodfrom $0.34/hr (opens in new tab)	24 GB · 14 GB free · ~78 tok/s · $1.21 / 1M tokens

Compare GPUs →