gpt-oss 20B
OpenAI's smaller open-weight model. Only a few billion parameters work on each token, so it runs far faster than its size suggests, and its reasoning is strong for a model that fits in 16 GB.
ollama run gpt-oss:20bNeeds about 16 GB of graphics memory, or unified memory on a Mac. First time? Install Ollama, then paste that into a terminal.
Download14 GB
Memory to runAbout 16 GB
Read per token2 GB (mixture of experts)
LicenseApache 2.0
Context128K tokens
ReleasedAugust 2025
Good for
- Reasoning, with an adjustable effort level
- Tool calling and agents
- Fast answers on 16 GB cards and Macs
Not the best pick for
- Image input: it is text only
- Creative writing, where it can sound flat
92 GPUs and Macs
What runs gpt-oss 20B
Estimated speed on each card at the smallest memory size that fits it, from memory bandwidth. As a mixture-of-experts model it reads only part of its weights per token, so it runs much faster than its size suggests.
Runs great 83 cards
- H200141 GB · ~840 tok/s
- H10080 GB · ~590 tok/s
- RTX 509032 GB · ~310 tok/s
- RTX Pro 6000 Blackwell96 GB · ~310 tok/s
- A10040 GB · ~270 tok/s
- RTX Pro 5000 Blackwell48 GB · ~240 tok/s
- Instinct MI10032 GB · ~220 tok/s
- Instinct MI5016 GB · ~180 tok/s
- RTX 3090 Ti24 GB · ~180 tok/s
- RTX 409024 GB · ~180 tok/s
- RTX 508016 GB · ~170 tok/s
- RTX 6000 Ada48 GB · ~170 tok/s
- RX 7900 XTX24 GB · ~170 tok/s
- RTX 309024 GB · ~160 tok/s
- A3024 GB · ~160 tok/s
- Tesla V10016 GB · ~160 tok/s
- RTX 5070 Ti16 GB · ~160 tok/s
- RTX 5080 Laptop16 GB · ~160 tok/s
- RTX 5090 Laptop24 GB · ~160 tok/s
- RTX Pro 4500 Blackwell32 GB · ~160 tok/s
- L40S48 GB · ~150 tok/s
- Radeon Pro W790048 GB · ~150 tok/s
- Apple M3 Ultra96 GB · ~140 tok/s
- RX 7900 XT20 GB · ~140 tok/s
- Apple M1 Ultra64 GB · ~140 tok/s
- Apple M2 Ultra64 GB · ~140 tok/s
- RTX A500024 GB · ~130 tok/s
- RTX A550024 GB · ~130 tok/s
- RTX A600048 GB · ~130 tok/s
- RTX 4080 Super16 GB · ~130 tok/s
- Tesla P10016 GB · ~130 tok/s
- RTX 408016 GB · ~130 tok/s
- A4048 GB · ~120 tok/s
- RTX 4070 Ti Super16 GB · ~120 tok/s
- Titan RTX24 GB · ~120 tok/s
- Quadro RTX 600024 GB · ~120 tok/s
- Quadro RTX 800048 GB · ~120 tok/s
- RTX Pro 4000 Blackwell24 GB · ~120 tok/s
- RTX A450020 GB · ~110 tok/s
- RX 907016 GB · ~110 tok/s
- RX 9070 XT16 GB · ~110 tok/s
- Radeon AI Pro R970032 GB · ~110 tok/s
- RX 7800 XT16 GB · ~110 tok/s
- A1024 GB · ~110 tok/s
- RTX 4090 Laptop16 GB · ~100 tok/s
- RTX 5000 Ada32 GB · ~100 tok/s
- RX 6950 XT16 GB · ~100 tok/s
- RX 7900 GRE16 GB · ~100 tok/s
- Radeon Pro W780032 GB · ~100 tok/s
- Arc A77016 GB · ~98 tok/s
- Apple M4 Max36 GB · ~96 tok/s
- RTX 3080 Ti Laptop16 GB · ~90 tok/s
- RX 680016 GB · ~90 tok/s
- RX 6800 XT16 GB · ~90 tok/s
- RX 6900 XT16 GB · ~90 tok/s
- Radeon Pro W680032 GB · ~90 tok/s
- Apple M5 Max36 GB · ~81 tok/s
- Arc Pro B6024 GB · ~80 tok/s
- RTX 5060 Ti16 GB · ~78 tok/s
- RTX 3080 Laptop16 GB · ~78 tok/s
- Quadro RTX 500016 GB · ~78 tok/s
- RTX A400016 GB · ~78 tok/s
- RTX 4500 Ada24 GB · ~76 tok/s
- Apple M1 Max32 GB · ~70 tok/s
- Apple M2 Max32 GB · ~70 tok/s
- Apple M3 Max36 GB · ~70 tok/s
- RTX 4000 Ada20 GB · ~63 tok/s
- Tesla P4024 GB · ~61 tok/s
- Tesla T416 GB · ~56 tok/s
- RX 9060 XT16 GB · ~56 tok/s
- Apple M5 Pro24 GB · ~54 tok/s
- L424 GB · ~53 tok/s
- RTX 4060 Ti16 GB · ~50 tok/s
- Tesla M4024 GB · ~50 tok/s
- RX 7600 XT16 GB · ~50 tok/s
- Apple M4 Pro24 GB · ~48 tok/s
- NVIDIA DGX Spark128 GB · ~48 tok/s
- Ryzen AI Max+ 395 (Radeon 8060S)64 GB · ~45 tok/s
- Ryzen AI Max 390 (Radeon 8050S)32 GB · ~45 tok/s
- RTX 2000 Ada16 GB · ~39 tok/s
- Jetson AGX Orin32 GB · ~36 tok/s
- Apple M1 Pro32 GB · ~35 tok/s
- Apple M2 Pro32 GB · ~35 tok/s
Usable 9 cards
Cheapest build from our list that runs it well: Used workstation + 2× Tesla P100, from $750 used, at about 130 tok/s. See the builds.
Learn
Read more about gpt-oss 20B
- A local coding assistant in VS Code: Continue, Cline and OllamaSet up a private AI coding assistant in VS Code with Ollama and Continue or Cline. Which model to pick for your card, and what to expect from it.
- AMD GPUs for local AI: ROCm, Vulkan and what works in practiceCan you run local LLMs on an AMD Radeon card? Which tools support it, ROCm vs Vulkan, the best AMD cards for AI, and the catches compared to NVIDIA.
- Build a local voice assistant: Whisper, a local LLM and text-to-speechHow a private voice assistant works on your own hardware, which parts to use for speech, thinking and speaking, and how to make it answer fast.
- gpt-oss 20B vs Qwen3 30B-A3B: two fast MoE models comparedOpenAI's gpt-oss 20B and Qwen3 30B-A3B both run fast on home hardware. Which fits your card, which is smarter where, and estimated speeds.
- How to increase Ollama's context length (num_ctx)Ollama's default context is short and it cuts off long prompts quietly. Three ways to raise num_ctx, and how much memory a longer context costs.
- How to judge a new local model in one eveningA new open model just dropped. A practical checklist to decide in a few hours whether it fits your hardware, works with your tools, and beats what you run now.
- Intel Arc for local AI: can the B580 and A770 run LLMs?How Intel Arc cards run local LLMs, which tools support them, estimated speeds for the B580 and A770 16 GB, and the software catches to know first.
- Local AI agents: tool calling, MCP and what works on small modelsHow agents and tool calling work with local models in Ollama, what MCP adds, which models handle tools reliably, and how to keep small models on track.
- Local AI for email: draft, summarize and sort mail privatelyUse a model on your own computer to draft replies, summarize threads and sort your inbox, without giving a cloud service your email. What works and what to avoid.
- Local AI for small business: a private assistant for your teamWhat a small business can do with AI running on its own hardware, what it costs, which build to buy, and how to share one machine with the whole team.
- Mixture of experts explained: why a 30B model can run like a 3BWhat MoE models are, what "active parameters" means, and why models like Qwen3 30B-A3B and gpt-oss are so fast on home hardware.
- Ollama commands cheat sheet: every command you'll actually useThe Ollama commands for running, downloading, listing, stopping and deleting models, plus the in-chat slash commands and the settings worth knowing.
- Ollama on a Mac: setup for Apple silicon, step by stepInstall Ollama on an M-series Mac, pick a model that fits your memory, check it's using the GPU, and change the settings that matter.
- Open model licenses explained: Apache 2.0, Llama and Gemma termsWhat the licenses on popular open models mean in practice for personal, business and product use. Apache 2.0, the Llama community license and Gemma terms.
- Run a local LLM on Windows: setup with Ollama and an NVIDIA or AMD GPUInstall Ollama on Windows 10 or 11, get it using your graphics card, change where models are stored, and fix the common problems.
- Run an LLM on CPU only: local AI on a laptop without a GPUWhich models run on a laptop with no graphics card, how fast to expect them, and the settings that make CPU-only local AI usable.
- Running gpt-oss 120B on a Mac: how much memory and how fastgpt-oss 120B needs about 67 GB for the GPU, so a 96 or 128 GB Mac. Which chips run it well, estimated speeds, and how to set it up.
- Running LLMs on a Mac: which M-series chip and how much memoryWhy Macs with Apple silicon are good at local AI, how much unified memory you need for each model size, and how they compare to a PC with a GPU.
- Serve a team on used GPUs: what our 90-card rack taught usHow to serve a local model to a team on cheap used GPUs. Our rack ran a billion tokens in 2.9 days on P100s, A4500s and mining cards. Builds and setup.
- Temperature, top-p and min-p: the sampling settings that matterWhat temperature and the other sampling settings do, which ones are worth changing for local models, and how to set them in Ollama.
- The best GPU for local AI under $1,000What to buy for local AI under $1,000 now that a used RTX 3090 costs more, how Tesla P100s and the RTX 3060 compare to new 16 GB cards and the RTX A4500, and what to buy for each kind of use.
- The best GPU for local AI under $300The best graphics cards for running local LLMs on a $300 budget, what each one runs, estimated speeds, and which to pick for easy setup or raw value.
- The best laptop for local AI: what to look for and what to avoidHow to pick a laptop for running local LLMs. Why laptop GPU names mislead, how much VRAM you need, estimated speeds, and when a MacBook is the better choice.
- The best local LLM for 16 GB of VRAM (4060 Ti, 5060 Ti, 5080)What to run on a 16 GB card like the RTX 4060 Ti 16 GB, 5060 Ti, 5080, RX 9070 XT or a used Tesla P100, with estimated speeds and what doesn't fit.
- The best local LLM for 24 GB of VRAM (RTX 3090, 4090, 7900 XTX)What to run on a 24 GB card, the community's favorite size for local AI. Picks for speed, smarts, images and code, with estimated speeds.
- The best local LLM for coding, by how much memory you haveWhich open models are worth using for code on 8, 12, 16, 24 GB and up, what they're good and bad at, and how to wire them into your editor.
- The best local LLM for reasoning and mathWhich open models reason best through math, logic and multi-step problems on 8 GB to 64 GB+, how thinking mode works, and why speed matters more than you'd expect.
- The best local model for agents and tool callingWhich open models call tools reliably on 12 GB to 64 GB+, how tool calling works with Ollama, and what to expect from local agents compared with cloud models.
- The Tesla P100 for local LLMs: 16 GB of HBM2 for under $100What we learned running a rack of 2016 data-center P100s for AI. Real speeds, what it takes to cool and power them, and the software catch.
- The Tesla P40 for local LLMs: 24 GB cheap, but is it fast enough?What the 24 GB Tesla P40 runs, how fast, how it compares to the P100, and the cooling, power and software catches before you buy one.
- Thinking models explained: reasoning mode, /think and effort levelsWhat reasoning mode does, when it's worth the wait, how to turn it on and off in Qwen3 and gpt-oss, and what it costs in speed and context.
- What is a system prompt, and how to set one in Ollama and other local toolsWhat a system prompt does, what to put in one, and how to set it in Ollama, a Modelfile, Open WebUI, LM Studio and the API.
- Where Ollama stores models, and how to change the locationThe default Ollama model folder on Windows, Mac and Linux, how to move it to another drive with OLLAMA_MODELS, and how to bring your existing models along.
Compare
Other models
- Qwen3 4BAbout 4 GB · Quick everyday chat and summaries
- Qwen3 8BAbout 7 GB · A solid all-rounder for chat and writing
- Gemma 3 12BAbout 10 GB · Reads images as well as text
- Qwen3 14BAbout 11 GB · Noticeably smarter at reasoning and code
- Gemma 3 27BAbout 19 GB · Google's best open model for its size
- Qwen3 30B-A3BAbout 21 GB · 30B smarts at small-model speed
- Qwen3 32BAbout 22 GB · Strong coding and long answers
- Llama 3.3 70BAbout 45 GB · Close to the big cloud models
- gpt-oss 120BAbout 67 GB · OpenAI's largest open model
- Qwen3 235B-A22BAbout 144 GB · Alibaba's flagship open model, for multi-GPU rigs
- DeepSeek R1 671BAbout 406 GB · The full DeepSeek R1, for the biggest rigs