We Value Your Privacy

We use cookies to enhance your browsing experience and analyze site traffic. All calculations happen locally in your browser - we never see or store your data. Learn more in our Privacy Policy

/
/
CalculateYogi
  1. Home
  2. Technology
  3. GPU Memory Calculator
Technology

GPU Memory Calculator

Pick your GPU and see which open-weight LLMs fit in its VRAM at your context length, which need quantizing, and how many cards the rest take.

Common rigs

Your hardware

Memory comes from each vendor's own product page. Pick the card you own, or enter a memory figure for anything not listed.

How you will run it

tokens
Did this calculator solve your problem today?

Contributor

Reviewed by

Last updated: August 23, 2026
SupportI build these free tools with love, late nights, and way too much coffee. If this calculator helped you, a small donation would mean the world to me and help keep this site running. Thank you for your kindness!

Related Calculators

You might also find these calculators useful

Model Size Calculator

Parameters, VRAM and GPU fit from a model's real architecture

AI Inference Cost Calculator

Compare self-hosted GPU vs API inference costs

Context Window Calculator

See whether your prompt fits an LLM context window

Fine-Tuning Cost Calculator

Price a fine-tuning run on published vendor rates

Which models will actually run on your GPU?

Most VRAM calculators start from a model and tell you how much memory it wants. This one starts from the card sitting in your machine. Pick it, say how long a context you need, and you get the whole open-weight catalogue sorted into what runs, what runs once quantized, and what would take more cards than you have. Memory comes from each vendor's own product page and every architecture is read from the model's own configuration file, so the parameter counts are derived rather than looked up.

What decides whether a model fits

Three things share your VRAM. The weights are fixed once you choose a precision: parameters multiplied by bytes per weight. The KV cache grows with context length, and it is the term people underestimate, because it scales with the key and value heads rather than the query heads. Grouped-query attention shrinks it dramatically, so a modern model at 32K tokens often needs less cache than an older one at 8K. Last comes the workspace the runtime needs for live tensors. On top of the three, allow around a fifth again for fragmentation and kernels, which is what stops a model that fits exactly on paper from loading at all.

What has to fit in VRAM

How to use this calculator

1

2

3

4

5

Common use cases

Deciding between two cards

Someone comparing a 16 GB card with a 24 GB one runs both through and sees exactly which models cross over, rather than guessing from a parameters-per-gigabyte rule.

Checking a rig before downloading

A developer about to pull a 40 GB checkpoint over a slow connection checks first whether it will load at all, and at which precision.

Sizing a rented instance

An engineer picking between an L40S and an A100 80GB compares what each one holds at the context length the product actually needs.

Finding the context ceiling

A team hitting out-of-memory errors at long context reads off the longest context their card can hold for that model, and decides whether to quantize or shorten the window.

Why start from the card

Buying and renting are card-first decisions

Nobody chooses a model and then goes shopping. You own a card, or you are choosing between two, and the real question is what each one unlocks. A 12 GB card and a 24 GB card differ by one quantization step in practice, not by twice the capability.

The quantization ladder is the answer more often than not

A model that overflows at BF16 usually fits at INT8, and almost always at INT4. Rather than making you try each setting, the table names the widest precision that fits on the hardware you described.

Context is the cost people forget

Weights are a fixed number you can look up. The cache is not, and at long context it can exceed the weights. The longest-context column tells you where your card actually runs out for each model.

Mixture-of-experts breaks the rules of thumb

A 235B mixture-of-experts model holds all 235 billion weights in memory but reads only around 22 billion per token. That makes it enormous to load and quick to run, and no per-billion rule of thumb captures both.

Frequently asked questions

At INT4, a 24 GB card comfortably holds a 32B dense model at a moderate context, and most 7B to 14B models with room to spare for a long window. At BF16 the same card tops out around 8B to 12B. Pick RTX 4090 above and the table gives the exact list rather than a rule of thumb, because the cache at your context length moves the boundary.

The speed ceiling needs a memory-bandwidth figure, and NVIDIA publishes one for every data-centre part and for the whole GeForce 50 series, but for nothing in the 30 or 40 series. The RTX 4090 product page prints the interface width and the memory type and stops there. Third-party sites quote a figure; it is not a vendor number, so this calculator leaves the column out rather than presenting someone else's arithmetic as a specification.

Card memory is sold in decimal gigabytes: a 24 GB card holds 24 billion bytes. Every figure on this page uses that same decimal gigabyte, so the requirement and the capacity can be subtracted directly. Most model tooling reports gibibytes instead, which are about seven per cent larger, so a model this page calls 20.4 GB shows as 19.0 GiB in llama.cpp or on a Hugging Face page. Same bytes, different unit — and mixing the two is exactly how a model that looked like it fit refuses to load.

Every expert has to be resident in VRAM, so the memory figure uses the total parameter count. Only the experts routed to each token are read during generation, so the speed ceiling uses the active count. That is why a 235B model with 22B active is both hard to fit and fast once it does.

Only roughly. The card count multiplies the memory available, which is right for a model split cleanly across cards, and the verdict column reports how many cards a model needs. It does not model interconnect overhead, per-card duplication of the cache, or the extra memory a tensor-parallel runtime holds, all of which make real multi-card setups slightly worse than the arithmetic.

Every architecture is transcribed from the model's own configuration file, or from the model's own technical paper where the repository is licence-gated, and the parameter count is derived from that architecture rather than taken from the name. A test checks each derivation against the size the model is named after. Card memory and bandwidth come from the vendor's product page.

CalculateYogi

The most comprehensive calculator web app. Free, fast, and accurate calculators for everyone.

Calculator Categories

  • Math
  • Finance
  • Health
  • Conversion
  • Date & Time
  • Statistics
  • Science
  • Engineering
  • Business
  • Everyday
  • Construction
  • Education
  • Technology
  • Food & Cooking
  • Sports
  • Climate & Environment
  • Agriculture & Ecology
  • Social Media
  • Other

Company

  • About
  • Contact
  • Contributors

Legal

  • Privacy Policy
  • Terms of Service
  • Editorial Policy

© 2026 CalculateYogi. All rights reserved.

Sitemap

Made with by the AppsYogi team