What are the hardware requirements for Oasst GPT Neox 20B 1000 Steps?

Oasst GPT Neox 20B 1000 Steps requires approximately 41.2 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 Oasst GPT Neox 20B 1000 Steps and how large is it?

Oasst GPT Neox 20B 1000 Steps is developed by dvruette, a model with 20b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Oasst GPT Neox 20B 1000 Steps perform on standard benchmarks?

Oasst GPT Neox 20B 1000 Steps has the following published scores: MMLU 26.39. Compare against reference models on this page or on the LLM Explorer leaderboards.

Oasst GPT Neox 20B 1000 Steps by dvruette — VRAM 41.2GB, 2K context

Name: Oasst GPT Neox 20B 1000 Steps
Author: dvruette

Oasst GPT Neox 20B 1000 Steps is an open-source language model by dvruette. Features: 20b LLM, VRAM: 41.2GB, Context: 2K, LLM Explorer Score: 0.1, Arc: 48.6, HellaSwag: 74.6, MMLU: 26.4, GSM8K: 3.1.

Endpoints compatible Gpt neox Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/dvruette/oasst-gpt-neox-20b-1000-steps

Oasst GPT Neox 20B 1000 Steps Parameters and Internals

Model Type

Conversational AI

Use Cases

Areas:

Research, Educational Tools, Conversational Agents

Applications:

Chatbots, Virtual Learning Assistants

Primary Use Cases:

Interactive Question Answering, Conversations with contextual understanding

Limitations:

Not suitable for decision-making in sensitive areas, Not guaranteed to be free of biases

Considerations:

Ensure proper oversight when deploying in high-stakes environments.

Additional Notes

Encourages community contribution and feedback for ongoing development.

Training Details

Data Sources:

Diverse internet sources, community feedback

Methodology:

Supervised fine-tuning

Model Architecture:

Modified transformer architecture optimized for dialogue.

Safety Evaluation

Methodologies:

Red-teaming, Adversarial testing

Findings:

The model shows improved safety mechanisms and consistent behavior in following guidelines.

Risk Categories:

Misinformation, Bias

Ethical Considerations:

The model is designed with consideration toward ethical usage, avoiding generating harmful or biased content.

Responsible Ai Considerations

Fairness:

The model adheres to fairness principles by attempting to avoid intrinsic biases present in training data.

Transparency:

OpenAssistant provides transparency reports and usage guidelines.

Accountability:

OpenAssistant is accountable for the model's outputs and continues to improve it based on community feedback.

Mitigation Strategies:

Implement ongoing model evaluations and updates based on community input.

Input Output

Input Format:

Text

Accepted Modalities:

Text

Output Format:

Text with contextual understanding

Performance Tips:

Ensure prompt adherence to guidelines to maintain the intention and context.

Release Notes

Version:

1.2

Date:

2023-08-15

Notes:

Added enhanced safety layers and improved accuracy in context retention.

LLM Name	Oasst GPT Neox 20B 1000 Steps
Repository 🤗	https://huggingface.co/dvruette/oasst-gpt-neox-20b-1000-steps
Model Size	20b
Required VRAM	41.2 GB
Updated	2026-07-29
Maintainer	dvruette
Model Type	gpt_neox
Model Files	10.0 GB: 1-of-5 10.0 GB: 2-of-5 10.0 GB: 3-of-5 10.0 GB: 4-of-5 1.2 GB: 5-of-5
Model Architecture	GPTNeoXForCausalLM
Context Length	2048
Model Max Length	2048
Transformers Version	4.28.0.dev0
Tokenizer Class	GPTNeoXTokenizer
Vocabulary Size	50288
Torch Data Type	bfloat16

Rank the Oasst GPT Neox 20B 1000 Steps 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 Oasst GPT Neox 20B 1000 Steps

Best Alternatives	Context / RAM	Downloads	Likes
GPT Neox 20B	2K / 40.8 GB	794665	583
GPT NeoXT Chat Base 20B	2K / 41.2 GB	256	694
EleutherAI GPT Neox 20B 4bits	2K / 12.5 GB	8	0
H2ogpt Oasst1 512 20B	2K / 41.2 GB	1064	39
...t Gm Oasst1 Multilang 1024 20B	2K / 41.2 GB	140	10
H2ogpt Gm Oasst1 En 1024 20B	2K / 41.2 GB	55	4
GPT Neox 20B Full Precision	2K / 82.5 GB	52	0
Oasst GPT Neox 20B 3000 Steps	2K / 41.2 GB	60	0
GPTNeoX 20B TestGen Dart V1.0	2K / 41.2 GB	5	2
GPT NeoX 20B Erebus	2K / 41.4 GB	576	88

Note: green Score (e.g. "73.2") means that the model is better than dvruette/oasst-gpt-neox-20b-1000-steps.