What are the hardware requirements for DeepSeek R1 0528 Quantized.w4a16?

DeepSeek R1 0528 Quantized.w4a16 requires approximately 263.8 GB of VRAM and supports a context window of 160K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed DeepSeek R1 0528 Quantized.w4a16 and how large is it?

DeepSeek R1 0528 Quantized.w4a16 is developed by RedHatAI, a model with 676.1b 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 DeepSeek R1 0528 Quantized.w4a16?

DeepSeek R1 0528 Quantized.w4a16 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

DeepSeek R1 0528 Quantized.w4a16 by RedHatAI — VRAM 263.8GB, 160K context

Name: DeepSeek R1 0528 Quantized.w4a16
Author: RedHatAI

DeepSeek R1 0528 Quantized.w4a16 is an open-source language model by RedHatAI. Features: 676.1b LLM, VRAM: 263.8GB, Context: 160K, License: mit, Quantized, LLM Explorer Score: 0.18.

Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-... Compressed-tensors Conversational Custom code Deepseek Deepseek v3 En Gptq Int4 Llmcompressor Neuralmagic Quantized Redhat Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/RedHatAI/DeepSeek-R1-0528-quantized.w4a16

DeepSeek R1 0528 Quantized.w4a16 Parameters and Internals

LLM Name	DeepSeek R1 0528 Quantized.w4a16
Repository 🤗	https://huggingface.co/RedHatAI/DeepSeek-R1-0528-quantized.w4a16
Base Model(s)	DeepSeek R1 0528 deepseek-ai/DeepSeek-R1-0528
Model Size	676.1b
Required VRAM	263.8 GB
Updated	2026-07-30
Maintainer	RedHatAI
Model Type	deepseek_v3
Model Files	1.9 GB: 1-of-63 1.2 GB: 2-of-63 1.2 GB: 3-of-63 1.2 GB: 4-of-63 6.3 GB: 5-of-63 6.3 GB: 6-of-63 6.3 GB: 7-of-63 6.3 GB: 8-of-63 6.3 GB: 9-of-63 6.3 GB: 10-of-63 6.3 GB: 11-of-63 6.3 GB: 12-of-63 6.3 GB: 13-of-63 6.3 GB: 14-of-63 6.3 GB: 15-of-63 6.3 GB: 16-of-63 6.3 GB: 17-of-63 6.3 GB: 18-of-63 6.3 GB: 19-of-63 6.3 GB: 20-of-63 6.3 GB: 21-of-63 6.3 GB: 22-of-63 6.3 GB: 23-of-63 6.3 GB: 24-of-63 6.3 GB: 25-of-63 6.3 GB: 26-of-63 6.3 GB: 27-of-63 6.3 GB: 28-of-63 6.3 GB: 29-of-63 6.3 GB: 30-of-63 6.3 GB: 31-of-63 6.3 GB: 32-of-63 6.3 GB: 33-of-63 6.3 GB: 34-of-63 6.3 GB: 35-of-63 6.3 GB: 36-of-63 6.3 GB: 37-of-63 6.3 GB: 38-of-63 6.3 GB: 39-of-63 6.3 GB: 40-of-63 6.3 GB: 41-of-63 6.3 GB: 42-of-63 6.3 GB: 43-of-63 6.3 GB: 44-of-63 6.3 GB: 45-of-63
Supported Languages	en
Quantization Type	gptq
Model Architecture	DeepseekV3ForCausalLM
License	mit
Context Length	163840
Model Max Length	163840
Transformers Version	4.52.0.dev0
Vocabulary Size	129280
Torch Data Type	bfloat16

Rank the DeepSeek R1 0528 Quantized.w4a16 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