mlx-optiq
← Models

gemma-4-e2b-it-qat-OptiQ-4bit

google/gemma-4-E2B-it-qat-q4_0-unquantized, quantized by OptiQ for Apple Silicon; takes images.

vision
On disk5.2 GB
Memoryabout 8 GB
Capability Score57.23
Imagesyes

Capability Score breakdown

The six metrics behind this model's Capability Score of 57.23, as reported in the model card. Values are compared with uniform 4-bit (QAT base).

  • This model
  • Uniform 4-bit (QAT base)
Capability Score breakdown: gemma-4-e2b-it-qat-OptiQ-4bitMMLU: This model 48.5, Uniform 4-bit (QAT base) 46.7; GSM8K: This model 58.6, Uniform 4-bit (QAT base) 56.2; IFEval: This model 66.0, Uniform 4-bit (QAT base) 67.7; BFCL-V3: This model 89.5, Uniform 4-bit (QAT base) 88.5; HumanEval: This model 62.8, Uniform 4-bit (QAT base) 59.8; HashHop: This model 18.0, Uniform 4-bit (QAT base) 12.0MMLU48.546.7GSM8K58.656.2IFEval66.067.7BFCL-V389.588.5HumanEval62.859.8HashHop18.012.0

Use it

Install

pip install mlx-optiq

Serve it

An OpenAI- and Anthropic-compatible server on this Mac. Every app below can use it.

optiq serve --model mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit

OptiQ Code

The terminal coding agent uses whatever model optiq serve is running.

optiq serve --model mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit
optiq code

OptiQ Lab

optiq lab --model mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit

Claude Code

optiq run claude -m mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit

Codex

optiq run codex -m mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit

OpenCode

optiq run opencode -m mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit

Mistral Vibe

optiq run vibe -m mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit

Hermes Agent

optiq run hermes -m mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit

OpenClaw

optiq run openclaw -m mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit

optiq run starts optiq serve if it is not running, sets the agent up for this session only, and stops the server when the agent exits.

mlx-lm, text only

from mlx_lm import load, generate

model, tokenizer = load("mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))

Details

FamilyGemma-4 · Gemma-4 docs
ArchitectureGemma 4
Base modelgoogle/gemma-4-E2B-it-qat-q4_0-unquantized
Size on disk5.2 GB
Memory neededabout 8 GB
Imagesyes, through optiq serve and the Lab
Capability Score57.23 (from the model card)
Downloads953 in the last month
Updated2026-10-18
Hugging Facemlx-community/gemma-4-e2b-it-qat-OptiQ-4bit