What are the hardware requirements for Granite 4.1 8B?

Granite 4.1 8B requires approximately 17.6 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 8B and how large is it?

Granite 4.1 8B is developed by ibm-granite, 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 4.1 8B perform on standard benchmarks?

Granite 4.1 8B has the following published scores: LMSYS ELO 1306. Compare against reference models on this page or on the LLM Explorer leaderboards.

Granite 4.1 8B by ibm-granite — VRAM 17.6GB, 128K context

Name: Granite 4.1 8B
Author: ibm-granite

Granite 4.1 8B is an open-source language model by ibm-granite. Features: 8b LLM, VRAM: 17.6GB, Context: 128K, License: apache-2.0, LLM Explorer Score: 0.46, ELO: 1306.

Arxiv:0000.00000 Conversational Deploy:azure Endpoints compatible Eval-results Granite Granite-4.1 Language Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/ibm-granite/granite-4.1-8b

Granite 4.1 8B Benchmarks

LMSys ELO 1306 vs 1272 (so35) 2.7%

^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 8B Parameters and Internals

LLM Name	Granite 4.1 8B
Repository 🤗	https://huggingface.co/ibm-granite/granite-4.1-8b
Model Size	8b
Required VRAM	17.6 GB
Updated	2026-05-07
Maintainer	ibm-granite
Model Type	granite
Model Files	4.8 GB: 1-of-4 5.0 GB: 2-of-4 5.0 GB: 3-of-4 2.8 GB: 4-of-4
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

Rank the Granite 4.1 8B 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

Quantized Models of the Granite 4.1 8B

Model	Likes	Downloads	VRAM
Granite 4.1 8B 8bit	2	876	9 GB
Granite 4.1 8B Fp8	13	17952	9 GB

Best Alternatives to Granite 4.1 8B

Best Alternatives	Context / RAM	Downloads	Likes
Granite 3.1 8B Instruct	128K / 16.4 GB	138888	168
Granite 4.1 8B Base	128K / 16.8 GB	11348	25
Granite Guardian 4.1 8B	128K / 16.8 GB	35637	33
Granite 3.1 8B Reasoning	128K / 16.4 GB	5	1
Granite 3.1 8B Instruct	128K / 16.4 GB	185	1
... Granite 4.1 8B Claude Fable 5	128K / 33.6 GB	846	5
Granite 3.2 8B Instruct	128K / 16.4 GB	3792	92
Granite 4.1 8B SFT Fable5	128K / 17.6 GB	237	0
...NG Coder V11.2800 Heretic BF16	128K / 16.8 GB	141	1
Granite 4.1 8B Nvfp4	128K / 4.7 GB	316	1