What are the hardware requirements for Granite 3B Code Instruct 2K?

Granite 3B Code Instruct 2K requires approximately 7 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Granite 3B Code Instruct 2K and how large is it?

Granite 3B Code Instruct 2K is developed by ibm-granite, a model with 3b 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 3B Code Instruct 2K?

Granite 3B Code Instruct 2K is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Granite 3B Code Instruct 2K by ibm-granite — VRAM 7GB, 2K context

Name: Granite 3B Code Instruct 2K
Author: ibm-granite

Granite 3B Code Instruct 2K is an open-source language model by ibm-granite. Features: 3b LLM, VRAM: 7GB, Context: 2K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.15.

Arxiv:2405.04324 Base model:finetune:ibm-granit... Base model:ibm-granite/granite... Code Conversational Dataset:bigcode/commitpackft Dataset:bugdaryan/sql-create-c... Dataset:garage-baind/open-plat... Dataset:glaive-function-callin... Dataset:glaiveai/glaive-code-a... Dataset:meta-math/metamathqa Dataset:nvidia/helpsteer Dataset:tiger-lab/mathinstruct Deploy:azure Granite Instruct Llama Model-index Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/ibm-granite/granite-3b-code-instruct-2k

Granite 3B Code Instruct 2K Parameters and Internals

Model Type

text-generation

Use Cases

Areas:

Code Assistance

Primary Use Cases:

Building coding assistants

Limitations:

Performance may be limited with out-of-domain programming languages

Considerations:

Providing few-shot examples can improve model performance in some cases.

Additional Notes

Finetuned using instruction datasets to enhance code instruction abilities.

Training Details

Data Sources:

CommitPackFT, MathInstruct, MetaMathQA, Glaive-Code-Assistant-v3, Glaive-Function-Calling-v2, NL2SQL11, Open-Platypus, HelpSteer

Methodology:

Fine-tuned for instruction following, logical reasoning, and problem-solving.

Hardware Used:

NVIDIA A100, NVIDIA H100

Responsible Ai Considerations

Mitigation Strategies:

Developers should perform safety testing and target-specific tuning before deploying these models.

LLM Name	Granite 3B Code Instruct 2K
Repository 🤗	https://huggingface.co/ibm-granite/granite-3b-code-instruct-2k
Base Model(s)	Granite 3B Code Base 2K ibm-granite/granite-3b-code-base-2k
Model Size	3b
Required VRAM	7 GB
Updated	2026-07-14
Maintainer	ibm-granite
Model Type	llama
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-2 2.0 GB: 2-of-2
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	4.41.0.dev0
Tokenizer Class	GPT2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	49152
Torch Data Type	bfloat16

Rank the Granite 3B Code Instruct 2K 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 3B Code Instruct 2K

Best Alternatives	Context / RAM	Downloads	Likes
Llama 3.2 3B Instruct	128K / 6.5 GB	1829572	2368
Llama 3.2 3B Instruct	128K / 6.5 GB	186207	96
Llama 3.2 3B Instruct	128K / 6.5 GB	991	9
Llama3.2 3B Manumit V1	128K / 6.4 GB	135	0
Alloma 3B Instruct	128K / 6.5 GB	55456	6
... V7 SFT 10voicebot Loopcleaned	128K / 6.5 GB	229	0
Llama 3.2 3B RP Toxic Fuse	128K / 6.4 GB	14	2
DeepSeek R1 Distill Llama 3B	128K / 6.5 GB	247	16
Schematron 3B	128K / 6.5 GB	969	336
ReasoningCore 3B T1 1	128K / 6.5 GB	8	1

Note: green Score (e.g. "73.2") means that the model is better than ibm-granite/granite-3b-code-instruct-2k.