How much VRAM does gpt-oss 120B (MoE) need?

gpt-oss 120B (MoE) requires approximately 0.0 GB VRAM at Q4_K_M quantization with 1K context and 10% overhead. At Q8_0 it needs ~0.0 GB, and at FP16 ~0.0 GB. Use the PCPARTGUIDE VRAM Calculator for exact requirements at any context length and overhead setting.

Can I run gpt-oss 120B (MoE) on a consumer GPU?

gpt-oss 120B (MoE) is a 117B parameter model and requires datacenter or multi-GPU setups. At Q4_K_M, it needs approximately 0.0 GB VRAM, which exceeds consumer GPU limits. It can run on H100 (80 GB) or A100 (80 GB) datacenter GPUs, or across multiple consumer GPUs with tensor parallelism.

What is gpt-oss 120B (MoE)'s context window?

gpt-oss 120B (MoE) supports a maximum context window of 131,072 tokens. The default context is 16,384 tokens. Larger contexts require proportionally more VRAM for the KV cache — use the VRAM Calculator to estimate exact requirements at your target context length.

What quantization level should I use for gpt-oss 120B (MoE)?

For gpt-oss 120B (MoE), start with Q4_K_M (~0.0 GB) if VRAM-limited. Q8_0 offers better quality at ~2× the VRAM. If you have enough VRAM, Q6_K (~1.5× Q4 VRAM) provides a good balance. For models this large, Q3_K_M may be necessary to fit on consumer hardware.

How does gpt-oss 120B (MoE) compare to similar models?

gpt-oss 120B (MoE) is a 117.0B MoE model from the OpenAI family. Browse other OpenAI models on the PCPARTGUIDE directory for comparable alternatives with different parameter counts and VRAM requirements. Use the Model Transparency scores to assess documentation quality and reproducibility.

OpenAIMoEApache 2.0

gpt-oss 120B (MoE)

117.0B

Parameters

5.1B

Active

128K

Max Context

MoE

Architecture

Aug 20, 2025

Released

Text

Modality

About gpt-oss 120B (MoE)

gpt-oss 120B is OpenAI's open-source MoE contribution — 117B total parameters with only 5.1B active per token. It uses 128 experts with top-4 routing and a shallow depth (36 layers) with wide hidden dimension (2880). Designed to fit on a single 80 GB GPU at moderate quantization with 128K YaRN context. The extreme sparsity (4.4% active) gives it speed comparable to a 5B dense model while delivering quality competitive with 70B+ models. A fascinating demonstration of how far MoE efficiency can be pushed.

Efficient MoEEnterpriseCodeReasoningCommercial

Technical Specifications

Total Parameters117.0B

Active Parameters5.1B per token

ArchitectureMixture of Experts

Total Experts128

Active Experts4 per token

Attention TypeGQA (Grouped Query Attention)

Hidden Dimensiond = 2,880

Transformer Layers36

Attention Heads64

KV Headsn_kv = 8

Head Dimensiond_head = 64

Activation FunctionSwiGLU

NormalizationRMSNorm

Position EmbeddingYaRN-extended RoPE

System Requirements

Estimated VRAM at 10% overhead for different quantization methods and context sizes.

Quantization	1K ctx	128K ctx
Q4_K_M0.50 B/W ~97% of FP16	60.55Datacenter GPU	69.48Datacenter GPU
Q8_01.00 B/W ~100% of FP16	121.0Cluster / Multi-GPU	130.0Cluster / Multi-GPU
F162.00 B/W Reference	242.0Cluster / Multi-GPU	250.9Cluster / Multi-GPU

Fits 24 GB consumer GPU

Fits 80 GB datacenter GPU

Requires cluster / multi-GPU

Other OpenAI Models

View All

gpt-oss 20B (MoE)

Params

21.0B

Layers

Context

128K

3.6B active

Find the right GPU for gpt-oss 120B (MoE)

Use the interactive VRAM Calculator to see exactly how much memory you need at any quantization level, context length, and overhead setting.

Open VRAM Calculator Best GPU Guide VRAM Guide