What are the hardware requirements for Llama 30B Instruct 2048?

Llama 30B Instruct 2048 requires approximately 65.2 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 Llama 30B Instruct 2048 and how large is it?

Llama 30B Instruct 2048 is developed by upstage, a model with 30b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Llama 30B Instruct 2048 perform on standard benchmarks?

Llama 30B Instruct 2048 has the following published scores: MMLU 61.9. Compare against reference models on this page or on the LLM Explorer leaderboards.

Llama 30B Instruct 2048 by upstage — VRAM 65.2GB, 2K context

Name: Llama 30B Instruct 2048
Author: upstage

Llama 30B Instruct 2048 is an open-source language model by upstage. Features: 30b LLM, VRAM: 65.2GB, Context: 2K, Instruction-Based, LLM Explorer Score: 0.1, Arc: 64.9, HellaSwag: 84.9, MMLU: 61.9, GSM8K: 17.8.

Dataset:gair/lima Dataset:metaeval/scienceqa tex... Dataset:open-orca/openorca Dataset:openbookqa Dataset:sciq Deploy:azure En Endpoints compatible Instruct Instruction Llama Pytorch Region:us Sharded Upstage

Model Card on HF 🤗: https://huggingface.co/upstage/llama-30b-instruct-2048

Llama 30B Instruct 2048 Parameters and Internals

Model Type

text generation

Supported Languages

en (native)

Training Details

Data Sources:

sciq, metaeval/ScienceQA_text_only, GAIR/lima, Open-Orca/OpenOrca, openbookqa

Context Length:

2048

Hardware Used:

A100x8 * 1

Input Output

Input Format:

up to 10k+ input tokens

Accepted Modalities:

text

Output Format:

text

Performance Tips:

Utilize `rope_scaling` option for longer inputs.

LLM Name	Llama 30B Instruct 2048
Repository 🤗	https://huggingface.co/upstage/llama-30b-instruct-2048
Model Size	30b
Required VRAM	65.2 GB
Updated	2026-07-09
Maintainer	upstage
Model Type	llama
Instruction-Based	Yes
Model Files	9.8 GB: 1-of-7 10.0 GB: 2-of-7 9.9 GB: 3-of-7 9.9 GB: 4-of-7 9.9 GB: 5-of-7 10.0 GB: 6-of-7 5.7 GB: 7-of-7
Supported Languages	en
Model Architecture	LlamaForCausalLM
Context Length	2048
Model Max Length	2048
Transformers Version	4.30.2
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Llama 30B Instruct 2048 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 Llama 30B Instruct 2048

Model	Likes	Downloads	VRAM
...e Llama 30B Instruct 2048 GGUF	4	865	13 GB
...ge Llama 30B Instruct 2048 AWQ	2	3	17 GB
...e Llama 30B Instruct 2048 GPTQ	22	27	16 GB

Best Alternatives to Llama 30B Instruct 2048

Best Alternatives	Context / RAM	Downloads	Likes
Tildeopen 30B Mu Instruct	64K / 60.9 GB	41	6
...ct 2048 Open Platypus 100steps	2K / 65.2 GB	88	0
...azarus Instruct PL Lora Unload	2K / 65.2 GB	85	0
Llama 30B Instruct	2K / 65.2 GB	333	24
...B 2048 Instruct PL Lora Unload	2K / 65.2 GB	97	1
...lama 30B Instruct 2048 PL Lora	2K / 65.2 GB	89	0
Open Instruct Human Mix 30B	2K / 130.5 GB	58	1
H2ogpt Oasst1 512 30B HF	2K / 65 GB	116	2
...pt Research Oig Oasst1 512 30B	2K / 81.1 GB	179	3
Veyra 30M Instruct	1K / 0.1 GB	151	3

Note: green Score (e.g. "73.2") means that the model is better than upstage/llama-30b-instruct-2048.