G

Gemma-4-31B-MoonGem

RP Chat

No ratings yet(0)
Prose
—(0)
Intelligence
—(0)
Knowledge
—(0)
Instruction Following
—(0)
Coding
—(0)
Steerability
—(0)
Sign in to vote

Ratings reflect this model as run on Arli AI.

View on Hugging FaceBack to Models

Hourly Usage

Performance Metrics

Typical Total Time

25.22s

Typical TTFT

6.54s

Typical Input TPS

1697.39

Typical Output TPS

16.01

Model Information

Context Size

262144

Quantization

r128 on INT8

Engine

vllm

Creation Method

Merge

Model Type

Gemma31B

Chat Template

Gemma4

Reasoning

Yes

Vision

Yes

Parameters

31B

Added At

9/26/2026

Use Gemma-4-31B-MoonGem with the Arli AI API

Gemma-4-31B-MoonGem is available through an OpenAI-compatible chat completions endpoint. Point any OpenAI SDK at https://api.arliai.com/v1, pass Gemma-4-31B-MoonGem as the model, and requests are served without logging.

cURL

curl https://api.arliai.com/v1/chat/completions \
  -H "Authorization: Bearer $ARLIAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Gemma-4-31B-MoonGem",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.arliai.com/v1",
    api_key="YOUR_ARLIAI_API_KEY",
)

response = client.chat.completions.create(
    model="Gemma-4-31B-MoonGem",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

See the text generation docs for every parameter, or read the quick start guide to make your first request.

Similar models

license: apache-2.0 tags:

  • merge
  • mergekit
  • roleplay model_creator: Ateron pipeline_tag: text-generation quantized_by: f16 language:
  • en base_model:
  • BirdToast/Gemma-4-31B-glimmer-rp-v0.1
  • nbeerbower/Gemma4-Gutenberg-31B
  • Jackrong/Gemopus-4-31B-it
  • bgg1996/Melinoe-Gemma4-31B-VL
  • AuriAetherwiing/G4-31B-Musica-v1
  • zerofata/G4-MeroMero-v2-31B
  • ReadyArt/gemma-4-31B-it-scotoma-2

🌑 MoonGem

MoonGem AI Persona

"..."

🌑 Overview

Moon.

🌑 Mergekit Configuration

Phase 1:

Bleed

ties Recipe 1 
merge_method: ties 
base_model: F:\AI\Merge\Gemma-4-Scotoma-V2
tokenizer_source: base
dtype: bfloat16
parameters:
  lambda: 1.0
models:
  - model: F:\AI\Merge\Gemma-4-Glimmer
    parameters:
      density: 0.15
      weight:
        - filter: "model.layers.5."
          value: 0.10
        - filter: "model.layers.11."
          value: 0.10
        - filter: "model.layers.17."
          value: 0.10
        - filter: "model.layers.23."
          value: 0.10
        - filter: "model.layers.29."
          value: 0.10
        - filter: "model.layers.35."
          value: 0.15
        - filter: "model.layers.41."
          value: 0.15
        - filter: "model.layers.47."
          value: 0.10
        - filter: "model.layers.53."
          value: 0.10
        - filter: "model.layers.59."
          value: 0.10
        - value: 0.10

  - model: F:\AI\Merge\Gemma-4-Gutenberg
    parameters:
      density: 0.15
      weight:
        - filter: "model.layers.5."
          value: 0.10
        - filter: "model.layers.11."
          value: 0.10
        - filter: "model.layers.17."
          value: 0.15
        - filter: "model.layers.23."
          value: 0.15
        - filter: "model.layers.29."
          value: 0.10
        - filter: "model.layers.35."
          value: 0.10
        - filter: "model.layers.41."
          value: 0.10
        - filter: "model.layers.47."
          value: 0.15
        - filter: "model.layers.53."
          value: 0.15
        - filter: "model.layers.59."
          value: 0.10
        - value: 0.10

  - model: F:\AI\Merge\Gemma-4-Gemopus
    parameters:
      density: 0.15
      weight:
        - filter: "model.layers.5."
          value: 0.15
        - filter: "model.layers.11."
          value: 0.15
        - filter: "model.layers.17."
          value: 0.10
        - filter: "model.layers.23."
          value: 0.10
        - filter: "model.layers.29."
          value: 0.10
        - filter: "model.layers.35."
          value: 0.15
        - filter: "model.layers.41."
          value: 0.10
        - filter: "model.layers.47."
          value: 0.10
        - filter: "model.layers.53."
          value: 0.10
        - filter: "model.layers.59."
          value: 0.10
        - value: 0.10

Phase 2:

Crimson

task_arithmetic Recipe 2 
merge_method: task_arithmetic
base_model: F:\AI\Merge\Gemma-4-Scotoma-V2
dtype: bfloat16
tokenizer_source: base
parameters:
  lambda: 1.0
models:
  - model: F:\AI\Merge\Gemma-4-Melinoe-VL
    parameters:
      weight: [0.10, 0.10, 0.15, 0.15, 0.10]

  - model: F:\AI\Merge\Gemma-4-MeroMero-V2
    parameters:
      weight: [0.10, 0.10, 0.15, 0.15, 0.10]

  - model: F:\AI\Merge\Gemma-4-MusicaV1
    parameters:
      weight: [0.10, 0.10, 0.15, 0.15, 0.10]

Phase 3:

Blood

task_arithmetic Recipe 3 
merge_method: task_arithmetic
base_model: F:\AI\Merge\Gemma-4-Scotoma-V2
dtype: bfloat16
tokenizer_source: base
parameters:
  lambda: 1.0
models:
  - model: F:\AI\Merge\Gemma-4-MicroMix-P1
    parameters:
      weight: 0.5

  - model: F:\AI\Merge\Gemma-4-MicroMix-P2
    parameters:
      weight: 0.5

🌑 Special Thanks

  • Google DeepMind: Gemma.
  • The Open-Source Community: People
  • Mergekit Fork: Zerofata.
  • To Nimbz: Cat.