What are the hardware requirements for Deepseek Coder 33B Base?

Deepseek Coder 33B Base requires approximately 66.5 GB of VRAM and supports a context window of 16K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Deepseek Coder 33B Base and how large is it?

Deepseek Coder 33B Base is developed by deepseek-ai, a model with 33b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Deepseek Coder 33B Base perform on standard benchmarks?

Deepseek Coder 33B Base has the following published scores: HumanEval (Python) 52.45. Compare against reference models on this page or on the LLM Explorer leaderboards.

Deepseek Coder 33B Base by deepseek-ai — VRAM 66.5GB, 16K context

Name: Deepseek Coder 33B Base
Author: deepseek-ai

Deepseek Coder 33B Base is an open-source language model by deepseek-ai. Features: 33b LLM, VRAM: 66.5GB, Context: 16K, License: other, Code Generating, LLM Explorer Score: 0.11, HumanEval: 52.5.

Codegen Deploy:azure Endpoints compatible Llama Pytorch Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/deepseek-ai/deepseek-coder-33b-base

Deepseek Coder 33B Base Parameters and Internals

Model Type

code generation, code completion

Use Cases

Areas:

Research, Commercial applications

Applications:

Project-level code completion, Infilling tasks

Primary Use Cases:

Code generation, Code completion, Repository-level code completion

Limitations:

Not specified

Supported Languages

English (High proficiency), Chinese (High proficiency)

Training Details

Data Sources:

Project-level code corpus

Data Volume:

2 trillion tokens

Methodology:

Grouped-Query Attention

Context Length:

16000

Input Output

Accepted Modalities:

Text

LLM Name	Deepseek Coder 33B Base
Repository 🤗	https://huggingface.co/deepseek-ai/deepseek-coder-33b-base
Model Size	33b
Required VRAM	66.5 GB
Updated	2026-07-21
Maintainer	deepseek-ai
Model Type	llama
Model Files	9.7 GB: 1-of-7 9.9 GB: 2-of-7 9.9 GB: 3-of-7 9.8 GB: 4-of-7 9.9 GB: 5-of-7 9.9 GB: 6-of-7 7.4 GB: 7-of-7 9.7 GB: 1-of-7 9.9 GB: 2-of-7 9.9 GB: 3-of-7 9.8 GB: 4-of-7 9.9 GB: 5-of-7 9.9 GB: 6-of-7 7.4 GB: 7-of-7
Generates Code	Yes
Model Architecture	LlamaForCausalLM
License	other
Context Length	16384
Model Max Length	16384
Transformers Version	4.34.1
Tokenizer Class	LlamaTokenizerFast
Beginning of Sentence Token	<｜begin▁of▁sentence｜>
End of Sentence Token	<｜end▁of▁sentence｜>
Vocabulary Size	32256
Torch Data Type	bfloat16

Rank the Deepseek Coder 33B Base 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 Deepseek Coder 33B Base

Model	Likes	Downloads	VRAM
Deepseek Coder 33B Base GGUF	9	3387	14 GB
Deepseek Coder 33B Base AWQ	4	22	18 GB
Deepseek Coder 33B Base GPTQ	2	18	17 GB

Best Alternatives to Deepseek Coder 33B Base

Best Alternatives	Context / RAM	Downloads	Likes
ReflectionCoder DS 33B	16K / 67 GB	8293	4
Deepseek Wizard 33B Slerp	16K / 35.3 GB	7	0
ValidateAI 3 33B Ties	16K / 66.5 GB	7	0
ValidateAI 2 33B AT	16K / 66.5 GB	6	0
WhiteRabbitNeo 33B V1	16K / 67 GB	214	95
Everyone Coder 33B Base	16K / 66.5 GB	86	21
Deepseek Coder 33B Instruct	16K / 66.5 GB	3376	580
Fortran2Cpp	16K / 67.3 GB	17	4
F2C Translator	16K / 67.3 GB	19	1
Llm4decompile 33B	16K / 66.5 GB	15	8

Note: green Score (e.g. "73.2") means that the model is better than deepseek-ai/deepseek-coder-33b-base.