gemma 3 12b it heretic

gemma

Open WeightModel weights are publicly available — can be downloaded and self-hosted

Released January 2026context N/A12b params

A mid-sized open-weight model built on Google's Gemma 3 12B foundation, fine-tuned by DreamFast under the Heretic label. It handles text-in, text-out tasks and ships in both GGUF and safetensors formats, making it straightforward to run locally. Specific capability adjustments from the fine-tuning process aren't well documented publicly.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Factual Knowledge

Moderate

Long Context

Moderate

Reasoning & Logic

Use Case Fit

Fit scores are AI-generated based on model capabilities, intended use, and technical specifications. Learn more

gemma 3 12b it heretic

gemma

Open WeightModel weights are publicly available — can be downloaded and self-hosted

Released January 2026context N/A12b params

A mid-sized open-weight model built on Google's Gemma 3 12B foundation, fine-tuned by DreamFast under the Heretic label. It handles text-in, text-out tasks and ships in both GGUF and safetensors formats, making it straightforward to run locally. Specific capability adjustments from the fine-tuning process aren't well documented publicly.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Factual Knowledge

Moderate

Long Context

Moderate

Reasoning & Logic

Use Case Fit

Fit scores are AI-generated based on model capabilities, intended use, and technical specifications. Learn more

Glossary

Fine-TunedA pre-trained model further trained on a smaller, task-specific dataset to improve performance on that task.Fine-TuningThe process of further training a pre-trained model on new data to adapt it for specific tasks or domains.GGUFA file format for quantized models designed for efficient CPU and GPU inference with llama.cpp.Open-Weight ModelA model whose trained weights are publicly released, allowing anyone to download and run it locally.SafetensorsA safe, fast file format for storing model weights, designed to prevent code execution vulnerabilities.Text-In, Text-OutA model that accepts text as input and produces text as output, without support for images, audio, or other data types.