Instructions to use Nesy1/g4_31b_v2_gembrain_equinox with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Nesy1/g4_31b_v2_gembrain_equinox with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Nesy1/g4_31b_v2_gembrain_equinox", dtype="auto")

llama-cpp-python

How to use Nesy1/g4_31b_v2_gembrain_equinox with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Nesy1/g4_31b_v2_gembrain_equinox",
	filename="g4_31b_v2_gembrain_equinox.Q4_E8K8VQ5FD5O6.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Nesy1/g4_31b_v2_gembrain_equinox with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

Use Docker

docker model run hf.co/Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

LM Studio
Jan
Ollama
How to use Nesy1/g4_31b_v2_gembrain_equinox with Ollama:
```
ollama run hf.co/Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M
```

Unsloth Studio new

How to use Nesy1/g4_31b_v2_gembrain_equinox with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Nesy1/g4_31b_v2_gembrain_equinox to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Nesy1/g4_31b_v2_gembrain_equinox to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Nesy1/g4_31b_v2_gembrain_equinox to start chatting

Pi new

How to use Nesy1/g4_31b_v2_gembrain_equinox with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Nesy1/g4_31b_v2_gembrain_equinox with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use Nesy1/g4_31b_v2_gembrain_equinox with Docker Model Runner:
```
docker model run hf.co/Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M
```

Lemonade

How to use Nesy1/g4_31b_v2_gembrain_equinox with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Nesy1/g4_31b_v2_gembrain_equinox:Q4_K_M

Run and chat with the model

lemonade run user.g4_31b_v2_gembrain_equinox-Q4_K_M

List all available models

lemonade list

Disclaimer

This is not my model, this is a thread full of quantizations of someone else's model. The aim with this thread is to give users access to beefy quants that are 20GB (approx) or under.

Overview:

This thread contains my custom quantizations of the model https://huggingface.co/rpDungeon/Gemma4-31b-Gembrain-Equinox. None of these are imatrixed, only static quants.

Model information below:

Gemma4-31b-Gembrain-Equinox

A V2 Fisher-protected community merge of two Gemma-4 31B creative-writing variants on top of stock google/gemma-4-31b-it. Built for the jaxxks / twisted / toasty research thread on instruct-preserving merges.

What's inside

Source	Role in the merge
`google/gemma-4-31b-it`	base / instruct anchor
Gembrain (community 31B)	style blender — reinforces prompt-attention
Equinox (community 31B)	unique style, neutral/realistic — best at low percentages

Recipe sketch: TIES-style merge of the two community deltas (Gembrain − IT, Equinox − IT) into IT, with Fisher importance + layer-importance damping applied to the combined delta to protect the high-importance instruct-following parameters. Scale 0.3.

Per jaxxks's characterization (2026-05-23): MeroMero would be more attention-to-detail-leaning; Gembrain reinforces user-prompt attention; Equinox carries a neutral/realistic finishing edge but is best at low percentages — the Fisher+layer damping is what lets us include Equinox at a non-trivial weight without losing instruct adherence.

Chat template — important

This repo ships a modified chat_template.jinja that prefills <|channel>thought\n on assistant turns when add_generation_prompt=True and enable_thinking=True. This patches a previously-observed "merges skip thinking" regression where the assistant would open a generation without entering the thought channel first.

enable_thinking=True → assistant turn begins inside the thought channel; produces a thought trace then folds back to the answer.
enable_thinking=False → behaves identically to stock Gemma-4 (empty thought-block convention preserved).

If you're loading via transformers, you'll pick this up automatically because chat_template.jinja is in the repo root. If you're loading via llama.cpp from a re-quantized GGUF, make sure the GGUF was built after the chat template was placed in the source dir, or apply it post-hoc via gguf_new_metadata.py --chat-template-file.

GGUF

A Q4_K_M GGUF of this exact merge (with the chat-template fix) is mirrored at:

ToastyPigeon/Gemma4-Test-GGUFs/g4_31b_v2_gembrain_equinox.Q4_K_M.gguf

Recommended llama-server flags for Gemma 4 31B at Q4_K_M on a dual-3090 box (subject to verification — see EVAL_GUIDELINES):

CUDA_VISIBLE_DEVICES=0,1 llama-server \
  --model g4_31b_v2_gembrain_equinox.Q4_K_M.gguf \
  --tensor-split 1,1 \
  --ctx-size 8192 --n-gpu-layers 999 \
  --parallel 8 --no-warmup --no-mmap --jinja \
  -fa on -ctk q8_0 -ctv q8_0

Client requests: cache_prompt: true. Do not pass --swa-full (default SWA-on is correct for Gemma 4).

Known caveats

This is a V2 merge. The V3 fisher+layer family (ToastyPigeon/g4-31b-v3-fisher-layer-test) supersedes it in our internal release thread but V2 remains an interesting "low-scale community blend on IT" data point.
Eval-side regression vs the local 31B-IT baseline was measured at roughly -19.78pp IFEval-strict in our first end-to-end test (2026-05-23). Most of that gap traced back to a chat-template-missing bug at the GGUF level (now fixed); a re-eval against the template-fixed GGUF is on the post-eval-opt queue.
The two source community models have their own licensing terms; check both before redistribution.

Reproduction notes (high level)

Pull stock google/gemma-4-31b-it weights.
Pull the two source community 31B variants (Gembrain, Equinox).
Compute Δ_GB = W_GB − W_IT, Δ_EQ = W_EQ − W_IT.
TIES-merge the two deltas (sign-conflict resolution + magnitude pruning).
Apply a Fisher + layer-importance mask to damp instruct-critical parameters, then add to IT at scale 0.3.
Copy chat_template.jinja (with the thought-channel prefill) into the merged dir before any quant step.
Save merged bf16 safetensors. Optionally convert to GGUF + quant.

For the masking technique itself, see INSTRUCT_MASKING_TECHNIQUES.md (cross-reference) or the project's instruct_mask/INSTRUCT_MASKING_TECHNIQUES.md.

Credits

Merge recipe: jaxxks (characterization) + twisted (mask infra) + toasty (training / orchestration)
Stock base: Google's Gemma 4 31B IT
Community variants: Gembrain, Equinox (originating authors retain credit for the source models)

Downloads last month: 424

GGUF

Model size

31B params

Architecture

gemma4

Hardware compatibility

4-bit

View +1 variant

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Nesy1/g4_31b_v2_gembrain_equinox

Base model

rpDungeon/Gemma4-31b-Gembrain-Equinox

Quantized

(4)

this model