What Models?

Pick the right model for your GPU — in seconds.

GPT-OSS 20B VRAM requirements: which GPUs can run it locally?

21B params 128K max context ReasoningTool use

The smallest GPU memory that runs GPT-OSS 20B well is 12 GB (Q8_0, e.g. GeForce RTX 5070). 86 of 132 GPUs run it well and 2 more are a tight fit. It's a mixture-of-experts model, so all weights must fit in memory but speed depends only on the active experts.

MMLU 75 SWE-bench Verified 60.7

VRAM needed per quantization

QuantizationWeights16K context32K contextFull 128K
Q4_K_M10.8 GB11.2 GB11.6 GB13.8 GB
Q8_011.3 GB11.6 GB12.0 GB14.2 GB

Weights plus fp16 KV cache. Leave ~1 GB headroom for the inference engine.

Run it locally

Each command downloads the GGUF from Hugging Face, then Ollama starts a chat and llama.cpp starts an OpenAI-compatible server with a web UI on port 8080.

Ollama ollama run hf.co/unsloth/gpt-oss-20b-GGUF:Q4_K_M
llama.cpp llama-server -hf unsloth/gpt-oss-20b-GGUF:Q4_K_M
Ollama ollama run hf.co/unsloth/gpt-oss-20b-GGUF:Q8_0
llama.cpp llama-server -hf unsloth/gpt-oss-20b-GGUF:Q8_0

Cheapest GPUs that can run GPT-OSS 20B

Lowest launch MSRP among GPUs that run its smallest quantization (Q4_K_M) well, with at least 16K context.

As an Amazon Associate, I earn from qualifying purchases.

GPUs that can run GPT-OSS 20B

Best quantization that fits on each GPU, with the context and speed it gets. For chips with several memory sizes, the smallest that works is shown.

NVIDIA

AMD

Intel

Apple

  • M1 16 GB Q8_0 128K context · ~21 tok/s
  • M1 Pro 16 GB Q8_0 128K context · ~56 tok/s
  • M2 16 GB Q8_0 128K context · ~30 tok/s
  • M2 Pro 16 GB Q8_0 128K context · ~56 tok/s
  • M3 16 GB Q8_0 128K context · ~30 tok/s
  • M4 16 GB Q8_0 128K context · ~36 tok/s
  • M5 16 GB Q8_0 128K context · ~44 tok/s
  • M6 16 GB Q8_0 128K context · ~44 tok/s
  • M3 Pro 18 GB Q8_0 128K context · ~44 tok/s
  • M4 Pro 24 GB Q8_0 128K context · ~73 tok/s
  • M5 Pro 24 GB Q8_0 128K context · ~80 tok/s
  • M1 Max 32 GB Q8_0 128K context · ~98 tok/s
  • M2 Max 32 GB Q8_0 128K context · ~98 tok/s
  • M3 Max 36 GB Q8_0 128K context · ~78 tok/s
  • M4 Max 36 GB Q8_0 128K context · ~100 tok/s
  • M5 Max 36 GB Q8_0 128K context · ~109 tok/s
  • M1 Ultra 64 GB Q8_0 128K context · ~125 tok/s
  • M2 Ultra 64 GB Q8_0 128K context · ~125 tok/s
  • M3 Ultra 96 GB Q8_0 128K context · ~127 tok/s
  • M5 Ultra 96 GB Q8_0 128K context · ~162 tok/s
Check your exact setup in the full tool →