What are the hardware requirements for Granite 20B Code Instruct?

Granite 20B Code Instruct requires approximately 40 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Granite 20B Code Instruct and how large is it?

Granite 20B Code Instruct is developed by ibm-granite, a model with 20b 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 20B Code Instruct?

Granite 20B Code Instruct is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Granite 20B Code Instruct by ibm-granite — VRAM 40GB

Name: Granite 20B Code Instruct
Author: ibm-granite

Granite 20B Code Instruct is an open-source language model by ibm-granite. Features: 20b LLM, VRAM: 40GB, License: apache-2.0, Instruction-Based, Code Generating, LLM Explorer Score: 0.16.

Arxiv:2405.04324 Autotrain compatible Base model:finetune:ibm-granit... Base model:ibm-granite/granite... Code Codegen 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 Endpoints compatible Gpt bigcode Granite Instruct Model-index Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/ibm-granite/granite-20b-code-instruct

Granite 20B Code Instruct Parameters and Internals

Model Type

text-generation

Use Cases

Areas:

coding assistants

Limitations:

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

Considerations:

Providing a few-shot example may help steer the model's output.

Supported Languages

Python (supported), JavaScript (supported), Java (supported), Go (supported), C++ (supported), Rust (supported), Other (limited)

Training Details

Data Sources:

bigcode/commitpackft, TIGER-Lab/MathInstruct, meta-math/MetaMathQA, glaiveai/glaive-code-assistant-v3, glaive-function-calling-v2, bugdaryan/sql-create-context-instruction, garage-bAInd/Open-Platypus, nvidia/HelpSteer

Methodology:

finetuned using instruction-response pairs

Hardware Used:

IBM's super computing clusters Vela and Blue Vela with NVIDIA A100 and H100 GPUs

Responsible Ai Considerations

Mitigation Strategies:

Developers should perform safety testing and target-specific tuning before deployment.

Release Notes

Date:

May 6th, 2024

LLM Name	Granite 20B Code Instruct
Repository 🤗	https://huggingface.co/ibm-granite/granite-20b-code-instruct
Base Model(s)	Granite 20B Code Base ibm-granite/granite-20b-code-base
Model Size	20b
Required VRAM	40 GB
Updated	2024-09-01
Maintainer	ibm-granite
Model Type	gpt_bigcode
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-9 4.9 GB: 2-of-9 4.9 GB: 3-of-9 4.9 GB: 4-of-9 4.9 GB: 5-of-9 4.9 GB: 6-of-9 4.9 GB: 7-of-9 4.9 GB: 8-of-9 0.7 GB: 9-of-9
Generates Code	Yes
Model Architecture	GPTBigCodeForCausalLM
License	apache-2.0
Model Max Length	8192
Transformers Version	4.41.0.dev0
Tokenizer Class	GPT2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	49152
Torch Data Type	bfloat16
Activation Function	gelu_pytorch_tanh

Rank the Granite 20B Code 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 20B Code Instruct

Best Alternatives	Context / RAM	Downloads	Likes
Granite 20B Code Instruct 8K	0K / 40 GB	9754	43
Granite 20B Code Instruct R1.1	0K / 40 GB	37	1

Note: green Score (e.g. "73.2") means that the model is better than ibm-granite/granite-20b-code-instruct.