Instructions to use srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4"

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 srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4

Run Hermes

hermes

MLX LM

How to use srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 / README.md

srv-sngh

Card: final complete benchmark table (base vs coder vs agentic, HE/HE+/MBPP/MBPP+/GSM8K, off+on)

00ca9ca verified 6 days ago

3.95 kB

license: gemma
library_name: mlx
base_model: yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF
base_model_relation: quantized
pipeline_tag: text-generation
model_type: gemma4_unified
tags:
  - nvfp4
  - mlx
  - krill
  - gemma4
  - gemma4_unified
  - apple-silicon
  - agentic
  - tool-use

gemma-4-12B-agentic-fable5-composer2.5-v2 — NVFP4 (MLX)

Original fine-tune by yuxinlu1 (gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2); this repo is an NVFP4 (MLX) requant for fast local inference on Apple Silicon.

A mixed-precision NVFP4 conversion of the Gemma‑4‑12B agentic fine‑tune (optimized for tool‑use / τ²‑bench). The weights are plain MLX safetensors; built for and tested with Krill — a pure Swift + MLX inference engine for Apple Silicon (no Python, no GGUF at inference). Loads in ~1.7 s at ~6.8 GB.

The format

Bulk weights are NVFP4 (4‑bit float, group_size 16); attention o_proj and the vision/audio projectors are kept at 8‑bit affine (group_size 64). This "protected" mixed recipe recovers the quality uniform 4‑bit loses on those sensitive modules while keeping 4‑bit speed and size.
Conversion: bf16 safetensors → key‑remap to MLX layout → NVFP4 requant. No GGUF round‑trip.

Compatibility (please read)

This is an MLX checkpoint — not GGUF, not a HF/transformers checkpoint. To load it an engine needs (1) the gemma4_unified architecture (text+vision+audio) and (2) the mixed‑precision NVFP4 config (top‑level nvfp4 + per‑module 8‑bit overrides). Today that means Krill; it is not drop‑in for vanilla mlx_lm/mlx_vlm, and not loadable by llama.cpp/Ollama (GGUF) or transformers/vLLM.

Install Krill & run

# Homebrew:
brew tap srvsngh99/krill && brew install krill
# …or one-line installer (Apple Silicon):
curl -fsSL https://raw.githubusercontent.com/srvsngh99/Krill/main/install.sh | sh

krill pull srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4   # by full path (alias TBD)
krill run  srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 "Write a Python LRU cache."
krill serve --model srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 --port 57455     # OpenAI-compatible API
KRILL_ENABLE_THINKING=1 krill run srv-sngh/gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4 "..."   # reasoning channel

Benchmarks

pass@1 / accuracy, single greedy pass, run with Krill on an M4 Pro. All three models are in this same NVFP4 format (true apples‑to‑apples). HumanEval+/MBPP+ are EvalPlus (stricter tests); MBPP = the 378‑problem EvalPlus set; GSM8K = 150‑problem subset, 8‑shot. Not EvalPlus‑leaderboard‑comparable.

Model	mode	HumanEval	HumanEval+	MBPP	MBPP+	GSM8K
Google gemma-4-12B-it (base)	off	57.3	56.7	42.1	37.6	95.3
Google gemma-4-12B-it (base)	on	48.8	48.8	49.5	43.9	90.7
coder v1	off	81.7	78.0	79.4	68.3	90.7
coder v1	on	80.5	76.2	80.4	68.8	90.0
agentic v2 ⟵ this model	off	83.5	81.7	84.1	74.1	90.7
agentic v2 ⟵ this model	on	86.0	82.9	83.6	73.0	91.3

Takeaways: the code/agentic fine‑tunes massively out‑code the Google base on HumanEval/MBPP, while the base is stronger at math (GSM8K). Reasoning‑on helps the fine‑tunes but tends to hurt the base's coding (it over‑reasons and mangles the code block). Decode ≈ 28 tok/s.

Credits & license

Fine‑tune © its original author (yuxinlu1 (gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2)); base model is Google Gemma 4, under the Gemma license. This repo only changes quantization/packaging.