How Much to Run AI
Know what your AI will actually cost to run.
What it does
"Can we self-host this model, and how much?" One number never ends the discussion. Pick a model (KIMI, GLM, DeepSeek…) and quantization, choose on-prem or cloud, set assumptions, and get a VRAM-matched hardware plan plus a monthly cost split. Every result shows its price sources and assumptions, refreshed on a regular schedule and never made up. Share one URL so the team debates the same inputs. A planning estimate, not a quote.
Estimate the GPU, deployment, electricity, and operations costs of running open models on your own hardware.
Your boss asks, "Should we run AI ourselves?" See not only how much it costs, but what the money is paying for. Pick a model, deployment option, and usage level to estimate the cost of running open models such as Kimi, GLM, DeepSeek, and Qwen on your own hardware. The result shows the VRAM, hardware, electricity, labor, and source assumptions behind the estimate, so you can check what went into it. VRAM ≈ params × bytes/param × runtime headroom (default ×1.2; adjustable for low/standard/high scenarios). FP16 = 2 bytes/param, INT4 = 0.5 bytes/param. For MoE models, all expert weights must reside in VRAM, so total parameter count is used. We start with the smallest GPU that fits the required…from howmuchtorunai.com
Does the same job
all alternatives →
- DMDeploy ML Models on a Budget2021 · github.com · ▲117
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, August 2026
the whole month →- TL
Life & fun · 10d ago · louisabraham.github.io
- SA
Hello HN! I found that picking out plausible but diverse skin tones for my digital art and game development projects was kind of difficult, and I got curious about if there was a way to define a color space that made it easy. I've built a color picker and procedural generation algorithm based on the space as well as a bunch of other fun js features and demos throughout the page that use the equations. If you find it interesting, I have lots of explanations of how I built it and what properties the space has. The methodology might be a bit shaky, but hopefully the result is as helpful for…
Life & fun · Aug 2026 · toneyalexander.github.io
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com