Gemma 4 E4B Instruct (MLX 4-bit) by Atomic Chat

The fastest GLM 5.2, Kimi K2.7 and MiniMax M3The fastest GLM 5.2, Kimi K2.7 and MiniMax M3 Discuss

Model details

Gemma 4 E4B Instruct (MLX 4-bit)

Atomic Chatgemma-4-E4B-it-MLX-4bitgemma

Open provider page Provider docs

Quick Info

Provider: Atomic Chat
Model key: gemma-4-E4B-it-MLX-4bit
Release date: Apr 2, 2026

Cost

A provider subscription or plan supersedes token-based pricing for this model.

Limits

Output tokens: 8,192 tokens
Context window: 32,768 tokens

Latest news about Gemma 4 E4B Instruct (MLX 4-bit)

Videos about Gemma 4 E4B Instruct (MLX 4-bit)

More models around Gemma 4 E4B Instruct (MLX 4-bit)

Browse family

19 providers

Gemma 4 31B IT

Currently listed through these providers:

Cerebras CrofAI Deep Infra FastRouter Friendli +14 more

12 providers

Gemma 4 26B A4B IT

Currently listed through these providers:

Cloudflare Workers AI Deep Infra evroc Google Hugging Face +7 more

4 providers

Gemma 3 27B

Currently listed through these providers:

NovitaAI OpenRouter Privatemode AI STACKIT

3 providers

Gemma 4 31B

Currently listed through these providers:

DigitalOcean NovitaAI OpenRouter

2 providers

Gemma 4 31B Instruct

Currently listed through these providers:

Berget.AI Together AI

2 providers

Gemma 4 26B A4B

Currently listed through these providers:

NovitaAI OpenRouter

2 providers

Google Gemma 3 27B Instruct

Currently listed through these providers:

Amazon Bedrock Venice AI

2 providers

Gemma SEA-LION v4 27B IT

Currently listed through these providers:

Cloudflare AI Gateway Cloudflare Workers AI

2 providers

Gemma 3 12B

Currently listed through these providers:

NovitaAI OpenRouter

2 providers

Google Gemma 3 12B

Currently listed through these providers:

Amazon Bedrock Helicone

1 provider

Gemma 4 Uncensored

Currently listed through:

Venice AI

1 provider

Pearl AI Gemma 4 31B Instruct

Currently listed through:

Together AI

Discuss this model

Powered byHyvor Talk

Add corrections, implementation notes, pricing changes, or usage caveats for other readers.