What are the hardware requirements for GPT Oss 120B BF16?

GPT Oss 120B BF16 requires approximately 142 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed GPT Oss 120B BF16 and how large is it?

GPT Oss 120B BF16 is developed by unsloth, a model with 120b 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 GPT Oss 120B BF16?

GPT Oss 120B BF16 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

GPT Oss 120B BF16 by unsloth — VRAM 142GB, 128K context

Name: GPT Oss 120B BF16
Author: unsloth

GPT Oss 120B BF16 is an open-source language model by unsloth. Features: 120b LLM, VRAM: 142GB, Context: 128K, License: apache-2.0, LLM Explorer Score: 0.24.

Base model:finetune:openai/gpt... Base model:openai/gpt-oss-120b Conversational Endpoints compatible Gpt oss Openai Region:us Safetensors Sharded Tensorflow Unsloth

Model Card on HF 🤗: https://huggingface.co/unsloth/gpt-oss-120b-BF16

GPT Oss 120B BF16 Parameters and Internals

LLM Name	GPT Oss 120B BF16
Repository 🤗	https://huggingface.co/unsloth/gpt-oss-120b-BF16
Base Model(s)	GPT Oss 120B openai/gpt-oss-120b
Model Size	120b
Required VRAM	142 GB
Updated	2026-07-21
Maintainer	unsloth
Model Type	gpt_oss
Model Files	1.2 GB: 1-of-73 4.2 GB: 2-of-73 2.2 GB: 3-of-73 4.2 GB: 4-of-73 2.2 GB: 5-of-73 4.2 GB: 6-of-73 2.2 GB: 7-of-73 4.2 GB: 8-of-73 2.2 GB: 9-of-73 4.2 GB: 10-of-73 2.2 GB: 11-of-73 4.2 GB: 12-of-73 2.2 GB: 13-of-73 4.2 GB: 14-of-73 2.2 GB: 15-of-73 4.2 GB: 16-of-73 2.2 GB: 17-of-73 4.2 GB: 18-of-73 2.2 GB: 19-of-73 4.2 GB: 20-of-73 2.2 GB: 21-of-73 4.2 GB: 22-of-73 2.2 GB: 23-of-73 4.2 GB: 24-of-73 2.2 GB: 25-of-73 4.2 GB: 26-of-73 2.2 GB: 27-of-73 4.2 GB: 28-of-73 2.2 GB: 29-of-73 4.2 GB: 30-of-73 2.2 GB: 31-of-73 4.2 GB: 32-of-73 2.2 GB: 33-of-73 4.2 GB: 34-of-73 2.2 GB: 35-of-73 4.2 GB: 36-of-73 2.2 GB: 37-of-73 4.2 GB: 38-of-73 2.2 GB: 39-of-73 4.2 GB: 40-of-73 2.2 GB: 41-of-73 4.2 GB: 42-of-73 2.2 GB: 43-of-73 4.2 GB: 44-of-73 2.2 GB: 45-of-73
Model Architecture	GptOssForCausalLM
License	apache-2.0
Context Length	131072
Model Max Length	131072
Transformers Version	4.56.0.dev0
Vocabulary Size	201088
Torch Data Type	bfloat16

Rank the GPT Oss 120B BF16 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 GPT Oss 120B BF16

Best Alternatives	Context / RAM	Downloads	Likes
GPT Oss 120B	128K / 65.1 GB	4232947	5068
GPT Oss 120B	128K / 65.1 GB	2742	22
GPT Oss 120B Mxfp4 Bf16	128K / 64.8 GB	1653	2
Medical GPT OSS Swallow 120B	128K / 139.8 GB	139	4
GPT Oss Safeguard 120B	128K / 65.1 GB	4032	99
Scal Lite 60B Code	128K / 34.8 GB	6	2
GPT Oss 120B Dequantized	128K / 135.6 GB	16	0
GPT Oss 120B MLX 8bit	128K / 123.7 GB	37801	13
GPT Oss 120B MXFP4 Q8	128K / 63.7 GB	7308	17
GPT Oss 120B Unsloth Bnb 4bit	128K / 62.1 GB	10211	17

Note: green Score (e.g. "73.2") means that the model is better than unsloth/gpt-oss-120b-BF16.