What are the hardware requirements for Openbuddy Deepseek 67B V15 Base?

Openbuddy Deepseek 67B V15 Base requires approximately 135 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Openbuddy Deepseek 67B V15 Base and how large is it?

Openbuddy Deepseek 67B V15 Base is developed by OpenBuddy, a model with 67b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Openbuddy Deepseek 67B V15 Base perform on standard benchmarks?

Openbuddy Deepseek 67B V15 Base has the following published scores: MMLU 70.97. Compare against reference models on this page or on the LLM Explorer leaderboards.

Openbuddy Deepseek 67B V15 Base by OpenBuddy — VRAM 135GB, 4K context

Name: Openbuddy Deepseek 67B V15 Base
Author: OpenBuddy

Openbuddy Deepseek 67B V15 Base is an open-source language model by OpenBuddy. Features: 67b LLM, VRAM: 135GB, Context: 4K, License: other, LLM Explorer Score: 0.11, Arc: 66.3, HellaSwag: 86, MMLU: 71, GSM8K: 56.9.

De En Fi Fr It Ja Ko Llama Region:us Ru Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/OpenBuddy/openbuddy-deepseek-67b-v15-base

Openbuddy Deepseek 67B V15 Base Parameters and Internals

Model Type

text-generation

Additional Notes

OpenBuddy models have inherent limitations and may potentially produce outputs that are erroneous, harmful, offensive, or otherwise undesirable.

Supported Languages

zh (Chinese), en (English), fr (French), de (German), ja (Japanese), ko (Korean), it (Italian), ru (Russian), fi (Finnish)

LLM Name	Openbuddy Deepseek 67B V15 Base
Repository 🤗	https://huggingface.co/OpenBuddy/openbuddy-deepseek-67b-v15-base
Model Size	67b
Required VRAM	135 GB
Updated	2026-08-08
Maintainer	OpenBuddy
Model Type	llama
Model Files	4.8 GB: 1-of-28 4.9 GB: 2-of-28 4.8 GB: 3-of-28 4.9 GB: 4-of-28 4.8 GB: 5-of-28 4.9 GB: 6-of-28 4.8 GB: 7-of-28 4.9 GB: 8-of-28 4.8 GB: 9-of-28 4.9 GB: 10-of-28 4.8 GB: 11-of-28 4.9 GB: 12-of-28 4.8 GB: 13-of-28 4.9 GB: 14-of-28 4.8 GB: 15-of-28 4.9 GB: 16-of-28 4.8 GB: 17-of-28 4.9 GB: 18-of-28 4.8 GB: 19-of-28 4.9 GB: 20-of-28 4.8 GB: 21-of-28 4.9 GB: 22-of-28 4.8 GB: 23-of-28 4.9 GB: 24-of-28 4.8 GB: 25-of-28 4.9 GB: 26-of-28 4.8 GB: 27-of-28 4.1 GB: 28-of-28
Supported Languages	zh en fr de ja ko it ru fi
Model Architecture	LlamaForCausalLM
License	other
Context Length	4096
Model Max Length	4096
Transformers Version	4.36.0.dev0
Tokenizer Class	LlamaTokenizerFast
Beginning of Sentence Token	<｜begin▁of▁sentence｜>
End of Sentence Token	<｜end▁of▁sentence｜>
Vocabulary Size	102400
Torch Data Type	bfloat16

Rank the Openbuddy Deepseek 67B V15 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 Openbuddy Deepseek 67B V15 Base

Model	Likes	Downloads	VRAM
...ddy Deepseek 67B V15 Base GGUF	3	1262	28 GB
...uddy Deepseek 67B V15 Base AWQ	0	7	37 GB
...ddy Deepseek 67B V15 Base GPTQ	0	5	36 GB

Best Alternatives to Openbuddy Deepseek 67B V15 Base

Best Alternatives	Context / RAM	Downloads	Likes
Deepseek Llm 67B Chat	4K / 135 GB	1034	207
...penbuddy Deepseek 67B V18.1 4K	4K / 135 GB	8526	1
Deepseek Llm 67B Base	4K / 135 GB	9029	130
...penbuddy Deepseek 67B V15.3 4K	4K / 135 GB	6	1
Openbuddy Deepseek 67B V15.2	4K / 135 GB	106	10
Openbuddy Deepseek 67B V15.1	4K / 135 GB	98	1
...penbuddy Deepseek 67B V15.3 4K	4K / 135 GB	11	1
Deepmoney 67B Chat	4K / 135 GB	20	23
DeepSeek 67B MMIQC	4K / 135 GB	9	1
...67B Spicy 3.1 1 5.0bpw H6 EXL2	4K / 43.7 GB	5	1

Note: green Score (e.g. "73.2") means that the model is better than OpenBuddy/openbuddy-deepseek-67b-v15-base.