Gemma-4-31B-GarnetV2

Creative model

View on Hugging FaceBack to Models

Hourly Usage

Performance Metrics

Avg. Total Time

55.86s

Avg. TTFT

3.27s

Avg. Prefill TPS

218.62

Avg. Gen TPS

32.92

Model Information

Context Size

262144

Quantization

r64

Engine

vllm

Creation Method

LoRA Finetune

Model Type

Gemma31B

Chat Template

Gemma4

Reasoning

Yes

Vision

Yes

Parameters

31B

Added At

5/1/2026

license: apache-2.0 base_model: google/gemma-4-31B-it pipeline_tag: text-generation datasets:

  • ConicCat/Gutenberg-SFT
  • ConicCat/Condor-SFT-Filtered

ConicCat/Gemma4-GarnetV2-31B

A finetune primarily focused on improving the prose and writing capabilities of Gemma 4. This does generalize strongly to roleplay and most other creative domains as well.

Features:

  • Improved longform writing capabilites; output context extension allows for prompting for up to 4000 words of text in one go.
  • Markedly less AI slop and identifiable Gemini-isms in writing.
  • Improved swipe or output diversity.
  • Fewer 'soft' refusals in writing.

Difference from V1

More / better roleplay data as well as shifting to using more primarily fantasy and sci fi books for training over literary fiction.

Datasets

  • internlm/Condor-SFT-20K for instruct; even though instruct capabilities are not the primary focus, adding some instruct data helps mitigate forgetting and maintains general intellect and instruction following capabilites.
  • ConicCat/Gutenberg-SFT. A reformatted version of the original Gutenberg DPO dataset by jondurbin for SFT with some slight augmentation to address many of the samples being overly long.
  • A dataset of backtranslated books. Unfortunately, I am unable to release this set as all of the data is under copyright.
  • A dash of a certain third owned archive.