What are the hardware requirements for Granite 3.1 8B Instruct?

Granite 3.1 8B Instruct requires approximately 16.4 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 3.1 8B Instruct and how large is it?

Granite 3.1 8B Instruct is developed by RedHatAI, a model with 8b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Granite 3.1 8B Instruct perform on standard benchmarks?

Granite 3.1 8B Instruct has the following published scores: LMSYS ELO 1208. Compare against reference models on this page or on the LLM Explorer leaderboards.

Granite 3.1 8B Instruct by RedHatAI — VRAM 16.4GB, 128K context

Name: Granite 3.1 8B Instruct
Author: RedHatAI

Granite 3.1 8B Instruct is an open-source language model by RedHatAI. Features: 8b LLM, VRAM: 16.4GB, Context: 128K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.3, ELO: 1208.

Arxiv:0000.00000 Ar Base model:finetune:ibm-granit... Base model:ibm-granite/granite... Conversational Cs De En Es Fr Granite Granite-3.1 Instruct It Ja Ko Language Nl Pt Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/RedHatAI/granite-3.1-8b-instruct

Granite 3.1 8B Instruct Benchmarks

LMSys ELO 1208 vs 1272 (so35) -5%

^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 3.1 8B Instruct Parameters and Internals

LLM Name	Granite 3.1 8B Instruct
Repository 🤗	https://huggingface.co/RedHatAI/granite-3.1-8b-instruct
Base Model(s)	ibm-granite/granite-3.1-8b-instruct ibm-granite/granite-3.1-8b-instruct
Model Size	8b
Required VRAM	16.4 GB
Updated	2026-06-29
Maintainer	RedHatAI
Model Type	granite
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-4 5.0 GB: 2-of-4 5.0 GB: 3-of-4 1.4 GB: 4-of-4
Supported Languages	en de es fr ja pt ar cs it ko nl zh
Model Architecture	GraniteForCausalLM
License	apache-2.0
Context Length	131072
Model Max Length	131072
Transformers Version	4.47.0
Tokenizer Class	GPT2Tokenizer
Padding Token	<\|end_of_text\|>
Vocabulary Size	49155
Torch Data Type	bfloat16
Errors	replace

Rank the Granite 3.1 8B Instruct 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 3.1 8B Instruct

Best Alternatives	Context / RAM	Downloads	Likes
Granite 3.1 8B Instruct	128K / 16.4 GB	138888	168
Granite 3.1 8B Reasoning	128K / 16.4 GB	5	1
Granite 3.2 8B Instruct	128K / 16.4 GB	3792	92
Granite 3.3 8B Instruct	128K / 16.4 GB	55103	160
Granite 3.3 8B Instruct FP8	128K / 8.4 GB	6980	3
Tema Q R19.0	128K / 17.6 GB	7	0
Granite 3.3 8B Math Prm V2	128K / 16.4 GB	203	14
...an 3.2 8B Factuality Detection	128K / 16.4 GB	138	4
Granite 3.3 8B Qiskit	128K / 16.4 GB	107	7
...ranite 3.2 8B Instruct Preview	128K / 16.4 GB	148	70

Note: green Score (e.g. "73.2") means that the model is better than RedHatAI/granite-3.1-8b-instruct.