Google

google/gemma-4-31B-it

Gemma 4 31B instruction-tuned serving with its optional MTP assistant.

Start this recipe View on Hugging Face

Model metadata

CategoryDenseParams31BContext262,144BackendsCUDA, METALTasksChat

Recipe authorship · Recorded configuration verification

Recipe authorship: Curated

Curated and generated describe how this recipe was written, not whether it ran successfully.

Recorded configuration verification

No current matching configuration verification record.

Records cover the tested default configuration on the named hardware. Matching release minor, variant, and rendering hash are required; changing options does not verify the new configuration.

Choose your actual device. Your browser cannot detect GPU memory.

Command builder

Hardware

Variant

mlx-community/gemma-4-31b-it-4bit

Mode: Server

Advanced

Mode

Context

Context length is measured in tokens. Use the slider or enter a number.

Slots

Features

Memory estimation: Why unavailable?

This feature is not available in the selected mode.

Estimated memory use

approx

Approximate memory allocation in GB. These estimates are not measured peak usage.

Weights: ≈ 16.2 GB
KV: ≈ 2.0 GB
Drafter: ≈ 0.0 GB
Runtime allowance: ≈ 1.7 GB

Estimated to leave ≈ 1.5 GB of the ≈ 21.4 GB available memory.

Next: prepare the executable

Notes

This tier is estimated from nearby measurements.

1. Install

Install in your terminal. Uncached model IDs download automatically on first use.

MLXCEL_INSTALL_DIR="${MLXCEL_INSTALL_DIR:-$HOME/.local/share/mlxcel}" \
  && mkdir -p "$MLXCEL_INSTALL_DIR" \
  && curl -fL -o "$MLXCEL_INSTALL_DIR/mlxcel.zip" https://github.com/lablup/mlxcel/releases/download/v0.8.1/mlxcel-macos-aarch64.zip \
  && unzip -o "$MLXCEL_INSTALL_DIR/mlxcel.zip" -d "$MLXCEL_INSTALL_DIR/mlxcel-macos-aarch64" \
  && export PATH="$MLXCEL_INSTALL_DIR/mlxcel-macos-aarch64:$PATH" \
  && mlxcel --version

Installation and download help

Run the install command in your terminal. The selected executable should then be available on PATH. If it is not found, repeat the displayed PATH export in the same terminal.

Model IDs download automatically when first used. The optional download command prepares the checkpoint ahead of time. Uncached models require internet access and enough disk space; gated models also require access approval and credentials.

2. Run

Run this command in the same terminal after preparation. For server modes, keep it running and wait for the listening/ready message. If memory is insufficient, select a smaller checkpoint or reduce context/concurrency in advanced settings.

Copy all includes the selected installation, checkpoint download, and displayed run commands, in that order. Run them in the same shell; multi-node commands belong on their respective machines.

mlxcel serve

This installation provides the full CLI. Use mlxcel serve; select a source build to use the standalone server.

mlxcel serve -m mlx-community/gemma-4-31b-it-4bit --alias gemma-4-31b-it \
  --ctx-size 8192 --n-parallel 1

Command flag help

-m: Selects the model or checkpoint repository.
--alias: Sets the served model name used by API clients.
--ctx-size: Sets the maximum context window for requests.
--n-parallel: Sets the number of parallel server slots.

3. Check the first result

For a server, use a second terminal after it is ready. A text chat request returns JSON with choices[0].message.content; specialized endpoints return their task output. Replace sample file paths with your own inputs where shown. Connection refused: check the server address/port and terminal logs. Model access/download errors: check the model ID, permissions, internet connection, and free disk space.

curl http://127.0.0.1:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"gemma-4-31b-it","messages":[{"role":"user","content":"Hello"}]}'

Performance evidence

No benchmark is recorded for this selected checkpoint and hardware combination. Commands and memory estimates use the catalog's model information.

Memory fit is a model-based estimate, separate from measured throughput and successful inference verification.

No measurement is available for this hardware yet. See the family page for related profiles. Family

Prerequisites

Install the mlxcel build for your selected hardware and check the selected checkpoint's memory estimate.
Allow internet access and enough disk space to download an uncached checkpoint; obtain Hugging Face access approval when the checkpoint requires it.

Recipe options

• The assistant is approximately 0.5B parameters and is included separately in the fit estimate.
• Speculative speedup depends on hardware and prompt shape; the target verifies every proposed token.

Family

Gemma 4

Dense

Sibling recipes

mlx-community/gemma-4-e2b-it-4bit

Guide

Gemma 4 31B IT

Pair the Gemma 4 target with its MTP assistant to propose draft tokens that the target model verifies.

Configure the recipe

Choose a memory tier that fits the 4-bit target and the assistant.
Start with 8192 context tokens and one parallel slot.
Enable spec decoding and select mtp.
Copy the download and run commands.

The target checkpoint is mlx-community/gemma-4-31b-it-4bit. Its assistant is mlx-community/gemma-4-31B-it-assistant-bf16, whose memory use appears in the drafter segment of the memory estimate.

Tune deliberately

Begin with four draft tokens and measure acceptance before raising the draft window.
The target verifies all accepted output, but speedup still varies by hardware and workload.
Tensor parallelism disables LoRA. Pipeline parallelism distributes target weights across machines, with the coordinator assigning the layers at runtime.

For lower-level speculative and distributed flags, see the mlxcel manual.