---
title: "Can this AI model run on my graphics card? Estimate model memory requirements and check whether a quantized LLM fits available GPU VRAM before loading or local inference."
method: POST
path: "/v1/model-memory-fit"
tags: ["ai", "llm", "model", "gpu", "vram", "memory", "fit", "fits", "quantized", "quantization", "inference", "graphics", "card"]
---

# Can this AI model run on my graphics card? Estimate model memory requirements and check whether a quantized LLM fits available GPU VRAM before loading or local inference.

`POST /v1/model-memory-fit`

## Request

- Base URL: `https://api.salt19.com`
- URL: `https://api.salt19.com/v1/model-memory-fit`
- Auth: none required

GPU VRAM and model memory-fit calculator for AI agents. Estimate whether quantized LLM or transformer model weights plus runtime overhead fit in available GPU memory before downloading, loading, deploying, or selecting a model. Useful for model selection, local inference planning, VRAM capacity checks, quantization decisions, and GPU deployment preflight.

## Headers

- `Idempotency-Key` string

## Request body

- object
  - `parameter_count_b` number, required
  - `quantization_bits` number, required
  - `vram_gb` number, required
  - `runtime_overhead_ratio` number

## Response `200`

Successful paid result

## Other responses

- `402`: x402 payment required

---

[API](https://skmtc.dev/salt19/apis/salt19-agent-utility-grid.md) · [All operations](https://skmtc.dev/salt19/apis/salt19-agent-utility-grid/llms.txt) · [OpenAPI document](https://skmtc.dev/salt19/apis/salt19-agent-utility-grid/revisions/c2520ca8ffbf?raw)