What are the hardware requirements for Granite 4.0 350M Base 8bit?

Granite 4.0 350M Base 8bit requires approximately 0.4 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Granite 4.0 350M Base 8bit and how large is it?

Granite 4.0 350M Base 8bit is developed by mlx-community, a model with 350m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Granite 4.0 350M Base 8bit?

Granite 4.0 350M Base 8bit is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Granite 4.0 350M Base 8bit by mlx-community — VRAM 0.4GB, 32K context

Name: Granite 4.0 350M Base 8bit
Author: mlx-community

Granite 4.0 350M Base 8bit is an open-source language model by mlx-community. Features: 350m LLM, VRAM: 0.4GB, Context: 32K, License: apache-2.0, Quantized, LLM Explorer Score: 0.21.

8-bit 8bit Base model:ibm-granite/granite... Base model:quantized:ibm-grani... Granite-4.0 Granitemoehybrid Language Mlx Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/mlx-community/granite-4.0-350m-base-8bit

Granite 4.0 350M Base 8bit Parameters and Internals

LLM Name	Granite 4.0 350M Base 8bit
Repository 🤗	https://huggingface.co/mlx-community/granite-4.0-350m-base-8bit
Base Model(s)	ibm-granite/granite-4.0-350m-base ibm-granite/granite-4.0-350m-base
Model Size	350m
Required VRAM	0.4 GB
Updated	2026-08-10
Maintainer	mlx-community
Model Type	granitemoehybrid
Model Files	0.4 GB
Quantization Type	8bit
Model Architecture	GraniteMoeHybridForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.56.0
Tokenizer Class	GPT2Tokenizer
Padding Token	<\|pad\|>
Vocabulary Size	100352
Torch Data Type	bfloat16

Rank the Granite 4.0 350M Base 8bit capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Granite 4.0 350M Base 8bit

Best Alternatives	Context / RAM	Downloads	Likes
...nite 4.0 350M Unsloth Bnb 4bit	32K / 0.4 GB	1743	0
...te 4.0 H 350M Unsloth Bnb 4bit	32K / 0.5 GB	232	1
Granite 4.0 H 350M Bnb 4bit	32K / 0.4 GB	15	1
Granite 4.0 350M	32K / 0.7 GB	20168	65
Granite 4.0 350M Base	32K / 0.7 GB	18470	26
Granite 4.0 H 350M	32K / 0.7 GB	7381	110
Granite 4.0 H 350M Base	32K / 0.7 GB	980	33
Granite 4.0 350M Gabliterated	32K / 0.7 GB	10	3
Granite 4.0 H 350M	32K / 0.7 GB	148	2
Granite 4.0 H 350M Base	32K / 0.7 GB	19	1

Note: green Score (e.g. "73.2") means that the model is better than mlx-community/granite-4.0-350m-base-8bit.