gemma-3-12b-it-heretic

Name: gemma-3-12b-it-heretic
Creator: p-e-w

p-e-w · Published release · Heretic · GGUF / Transformers / MLX

Model familyGemma

Parameters12B

Estimated Q4 VRAMAbout 7.9 GB

30-day downloads7.3K

Declared licensegemma

Source repositories5

Model source

Published by p-e-w on Hugging Face. This index organizes public repository metadata for discovery and comparison. Refer to the publisher and independent testing for actual model behavior.

Declared base model

google/gemma-3-12b-pt ↗

Hardware and context

Estimated at 7.9 GB VRAM for 12B parameters with 4-bit weights, including 15% weight overhead and 1 GB of baseline headroom. KV cache, context length, multimodal components and concurrency increase actual memory usage.The parameter count was parsed from the model name and has not been independently verified.

Context length: Not stated in public metadata. Please consult the model card.

Download sources and quantized variants

5 repositories

HUGGING FACE REPOSITORY	Format	30-day downloads	Declared license
mradermacher/gemma-3-12b-it-heretic-GGUF ↗	GGUF	5.4K	gemma
p-e-w/gemma-3-12b-it-heretic ↗	Transformers	907	gemma
mradermacher/gemma-3-12b-it-heretic-i1-GGUF ↗	GGUF	662	gemma
mfielding92/littlemonster-reasoning-12B-QKVO-heretic-GGUF ↗	GGUF	229	apache-2.0
CallMcMargin/gemma-3-12b-it-heretic-mlx-bf16-affine-qgroup32-q8 ↗	MLX	61	gemma

Uncensored is a publisher description, not a guarantee of zero refusals, capability, accuracy or safety. This index does not measure refusal rates or performance benchmarks.

gemma-3-12b-it-heretic

Model source

Declared base model

Hardware and context

Download sources and quantized variants

More from this family

Gemma-4-E4B-Uncensored-HauhauCS-Aggressive

Gemma4-12B-QAT-Uncensored-HauhauCS-Balanced

Gemma4-26B-A4B-QAT-Uncensored-HauhauCS-Balanced-MTP

gemma-4-31B-it-abliterated