BestLLMfor EN Your hardware. Your LLM. Your call.
APIOpen data Find my LLM
Model fiche

gpt-oss 20B

By OpenAI · United States

Updated 2026-07-13

chat general reasoning moe small
Parameters
21B
License
Apache 2.0
Context
125k
VRAM (Q4)
13 GB
Released
April 2025

Overview

OpenAI's compact open-weight MoE with 3.6B active out of 21B total parameters. Matches o3-mini on a laptop-class GPU under Apache 2.0.

When to pick this model

  • Local development on consumer or workstation GPUs
  • Edge deployments needing frontier-vendor quality
  • 128k-context tasks without datacenter hardware
  • Apache-licensed replacement for o3-mini API calls

VRAM requirements by quantization

VRAM REQUIRED (GB)812162432Q4_K_M13 GBQ5_K_M16 GBQ8_023 GBFP1642 GB
QuantizationVRAM required
Q4_K_M (recommended)13 GB
Q5_K_M16 GB
Q8_023 GB
FP16 (no quantization)42 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

In practice, gpt-oss 20B needs a 16 GB card at Q4_K_M (13 GB). Stepping up to Q8_0 nearly doubles the footprint to 23 GB, and unquantized FP16 weights take 42 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.

Without a GPU, gpt-oss 20B needs roughly 18 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 20 tokens/sec on entry-level GPUs, on the order of 55 tokens/sec on a mid-range card, and up to 130 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.

What hardware do you need

The table below matches gpt-oss 20B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.

GPU memoryExample cardsBest fit for gpt-oss 20B
8 GBRTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GBDoes not fit — needs 13 GB at Q4_K_M
12 GBRTX 5070, RTX 5070 Ti Laptop, RTX 4080 LaptopDoes not fit — needs 13 GB at Q4_K_M
16 GBRTX 5080, RTX 4080 Super, Radeon RX 9070 XTQ5_K_M (16 GB used)
24 GBRTX 4090, Radeon RX 7900 XTX, RTX 5090 LaptopQ8_0 (23 GB used)
32 GBRTX 5090Q8_0 (23 GB used)

Which GPU should you buy to run gpt-oss 20B?

To run gpt-oss 20B locally at Q4, you need ~13 GB of VRAM. The best value for this is a RTX 5070 Ti (16 GB VRAM).

Check RTX 5070 Ti price on Amazon →

As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.

Strengths

  • Apache 2.0 with full commercial freedom
  • Around 13 GB VRAM at Q4 — runs on a 16 GB card
  • OpenAI quality in an accessible footprint
  • Native 128k context

Limitations

  • MoE format uses more VRAM than equivalent dense models
  • Fewer community fine-tunes than Llama or Qwen

Typical workloads

In our catalog grid, gpt-oss 20B is filed under Edge reasoning, Tool use, Laptop — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: multi-step reasoning and math-flavoured tasks.

The 125k-token context window covers long chats and mid-sized documents, though very large retrieval workloads will need chunking. The Apache 2.0 license is permissive, so shipping it inside a commercial product raises no special legal questions.

Architecture & training

Architecture: MoE · ~21B total / ~4B active · OpenAI open-source compact

Training: Lightweight version of OpenAI's open-source GPT series, ideal for local deployment.

Verdict

The clear default for local OpenAI-quality inference — accessible VRAM, 128k context, and a real license.

Quick start

ollama run openai/gpt-oss:20b

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.

Similar models worth comparing

Frequently asked questions

How much VRAM does gpt-oss 20B need?

At the recommended Q4_K_M quantization, gpt-oss 20B needs about 13 GB of VRAM. Q8_0 takes 23 GB, and unquantized FP16 weights take 42 GB.

Can gpt-oss 20B run without a GPU?

Yes — with roughly 18 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.

What context window does gpt-oss 20B support?

gpt-oss 20B supports a 125k-token context window (128,000 tokens).

Can I use gpt-oss 20B commercially?

Yes. gpt-oss 20B is released under Apache 2.0, a permissive open-source license that allows commercial use, modification and redistribution.

How fast is gpt-oss 20B on consumer hardware?

Our compatibility engine estimates on the order of 55 tokens/sec on a mid-range GPU and up to 130 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.

Which quantization of gpt-oss 20B should I download first?

Start with Q4_K_M (13 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. On a 24 GB card you can run up to Q8_0.

Tools

Is gpt-oss 20B the right pick for you?

Compute self-hosted ROI → Back to catalog