Local inference, no cloud
Which hardware do you need to run your LLM locally?
Give a budget, a model, or both. You get three concrete setups: enough memory or not, an estimated speed, and a price we label as an estimate.
11 models · 19 setups
Common questions
- Does Llama 70B fit on an RTX 4090?
- Not in full at Q4. The estimate sits around 40 GB, and a 4090 has 24 GB. Two 24 GB cards, or a Mac with enough unified memory, can. That is still an estimate, not a measured GGUF.
- Are the tokens per second measured?
- No. Tokens per second are estimated from memory bandwidth. There is no benchmark source: benchmarkSourceUrl stays null.
- Are the prices accurate?
- They are indicative for 2026-10-14, incl. VAT, for the EU (BE/NL). There is no shop source: sourceUrl stays null.
- Does the newsletter store my address?
- Only when RESEND_API_KEY and RESEND_AUDIENCE_ID are set. No email is sent. Without those variables nothing is stored.