What are the hardware requirements for Upstage Llama1 65B Instruct GPTQ?

Upstage Llama1 65B Instruct GPTQ requires approximately 34.7 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 Upstage Llama1 65B Instruct GPTQ and how large is it?

Upstage Llama1 65B Instruct GPTQ is developed by TheBloke, a model with 65b 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 Upstage Llama1 65B Instruct GPTQ?

Upstage Llama1 65B Instruct GPTQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Upstage Llama1 65B Instruct GPTQ by TheBloke — VRAM 34.7GB, 2K context

Name: Upstage Llama1 65B Instruct GPTQ
Author: TheBloke

Upstage Llama1 65B Instruct GPTQ is an open-source language model by TheBloke. Features: 65b LLM, VRAM: 34.7GB, Context: 2K, License: other, Quantized, Instruction-Based, LLM Explorer Score: 0.07.

4-bit Base model:quantized:upstage/l... Base model:upstage/llama-65b-i... En Gptq Instruct Instruction Llama Quantized Region:us Safetensors Upstage

Model Card on HF 🤗: https://huggingface.co/TheBloke/Upstage-Llama1-65B-Instruct-GPTQ

Upstage Llama1 65B Instruct GPTQ Parameters and Internals

Model Type

llama

Use Cases

Considerations:

Handle up to 10k+ input tokens,

Additional Notes

Supports up to 10k+ input tokens with rope_scaling option.

Supported Languages

English (fluent)

Training Details

Data Sources:

Orca-style dataset

Methodology:

fine-tuned

Hardware Used:

A100x8 * 4

Model Architecture:

dynamic with rope_scaling

Input Output

Input Format:

### System: {system_message} ### User: {prompt} ### Assistant:

Accepted Modalities:

text

Performance Tips:

Using quantisation with Act Order can improve accuracy with lower VRAM requirements.

LLM Name	Upstage Llama1 65B Instruct GPTQ
Repository 🤗	https://huggingface.co/TheBloke/Upstage-Llama1-65B-Instruct-GPTQ
Model Name	Llama 65B Instruct
Model Creator	upstage
Base Model(s)	Llama 65B Instruct upstage/llama-65b-instruct
Model Size	65b
Required VRAM	34.7 GB
Updated	2026-07-14
Maintainer	TheBloke
Model Type	llama
Instruction-Based	Yes
Model Files	34.7 GB
Supported Languages	en
GPTQ Quantization	Yes
Quantization Type	gptq
Model Architecture	LlamaForCausalLM
License	other
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 Upstage Llama1 65B Instruct GPTQ 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 Upstage Llama1 65B Instruct GPTQ

Best Alternatives	Context / RAM	Downloads	Likes
...en Instruct Human Mix 65B GPTQ	2K / 33.5 GB	11	0
Lite Oute 1 65M Instruct	2K / 0.3 GB	15	12
...pen Instruct Human Mix 65B AWQ	2K / 34.7 GB	13	0
Llama 65B Instruct	2K / 130.4 GB	198	15
...pstage Llama1 65B Instruct AWQ	2K / 34.7 GB	9	1
Open Instruct Human Mix 65B	2K / 260.8 GB	173	4
Open Instruct Sharegpt 65B	2K / 260.8 GB	45	2

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/Upstage-Llama1-65B-Instruct-GPTQ.