What are the hardware requirements for Granite 4.1 30B?

Granite 4.1 30B requires approximately 57.8 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Granite 4.1 30B and how large is it?

Granite 4.1 30B is developed by ibm-granite, a model with 30b 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 4.1 30B?

Granite 4.1 30B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Granite 4.1 30B by ibm-granite — VRAM 57.8GB, 128K context

Name: Granite 4.1 30B
Author: ibm-granite

Granite 4.1 30B is an open-source language model by ibm-granite. Features: 30b LLM, VRAM: 57.8GB, Context: 128K, License: apache-2.0, LLM Explorer Score: 0.3.

Arxiv:0000.00000 Conversational Endpoints compatible Granite Granite-4.0 Language Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/ibm-granite/granite-4.1-30b

Granite 4.1 30B Benchmarks

LLME Score: 0.29911

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Granite 4.1 30B (ibm-granite/granite-4.1-30b)

🌟 Advertise your project 🚀

Granite 4.1 30B Parameters and Internals

LLM Name	Granite 4.1 30B
Repository 🤗	https://huggingface.co/ibm-granite/granite-4.1-30b
Model Size	30b
Required VRAM	57.8 GB
Updated	2026-04-29
Maintainer	ibm-granite
Model Type	granite
Model Files	5.0 GB: 1-of-12 4.8 GB: 2-of-12 5.0 GB: 3-of-12 4.8 GB: 4-of-12 5.0 GB: 5-of-12 4.8 GB: 6-of-12 5.0 GB: 7-of-12 4.8 GB: 8-of-12 5.0 GB: 9-of-12 4.8 GB: 10-of-12 5.0 GB: 11-of-12 3.8 GB: 12-of-12
Model Architecture	GraniteForCausalLM
License	apache-2.0
Context Length	131072
Model Max Length	131072
Transformers Version	4.53.3
Tokenizer Class	GPT2Tokenizer
Padding Token	<\|pad\|>
Vocabulary Size	100352
Torch Data Type	bfloat16

Quantized Models of the Granite 4.1 30B

Model	Likes	Downloads	VRAM
Granite 4.1 30B Fp8	8	3824	30 GB
Granite 4.1 30B 4bit	2	476	18 GB

Best Alternatives to Granite 4.1 30B

Best Alternatives	Context / RAM	Downloads	Likes
...30B Claude 4.6 Opus Thinking X	128K / 58.6 GB	8	1
Granite 4.1 30B Base	128K / 57.8 GB	1280	24
Granite 4.1 30B Mxfp8	128K / 29.8 GB	660	2
Granite 4.1 30B	128K / 57.8 GB	287	0
...6 Opus Thinking Charles Xavier	128K / 58.6 GB	47	0
Unsloth Granite 4.1 30B Mlx	128K / 57.4 GB	49	0
...ui Granite 4.1 30B Abliterated	128K / 57.8 GB	162	9
Granite 4.1 30B 4bit	128K / 18.1 GB	476	2
Granite 4.1 30B Fp8	128K / 30.1 GB	3824	8

Note: green Score (e.g. "73.2") means that the model is better than ibm-granite/granite-4.1-30b.

Rank the Granite 4.1 30B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

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

What open-source LLMs or SLMs are you in search of? 53666 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260328a

Support LLM Explorer

Granite 4.1 30B by ibm-granite