Modal GPU Cost Calculator
Pick a Modal GPU (B300, B200, H200, H100, RTX PRO 6000, A100 80/40GB, L40S, A10, L4, T4), enter GPU seconds per request, requests per month, CPU cores and memory, container starts with idle keep-alive seconds, and a plan. The calculator returns cost per request, per 1,000 requests, monthly total and the equivalent $/GPU-hour. Estimates only; modal.com/pricing prevails.
Math runs in your browser. Nothing is uploaded.
Monthly total
—
Per request
—
Breakdown
—
Formula: billed seconds = GPU seconds per request × requests + starts × idle keep-alive seconds (Modal keeps a container warm after its last input and bills that idle time). GPU = billed s × $/s; CPU = billed s × max(cores, 0.125) × $0.0000131 (preemptible Functions; 3× for non-preemptible / Sandboxes); memory = billed s × GiB × $0.00000222. Total = compute − optional included credit + plan fee ($0 Starter incl. $30; $250 Team incl. $100). Official check: Stable Diffusion on A10 at ~1.5 s/image with 1 core + 4 GiB ≈ $0.00049 per image.
Estimates only. Official list prices as of 2026-11-01 (Asia/Shanghai); the provider's pricing page prevails. Sources: Modal pricing
RunPod GPU Cost Calculator · GPU Cloud Cost Comparison · fal.ai Cost Calculator · Together AI Token Cost Calculator · Fireworks Token Cost Calculator · DeepInfra Token Cost Calculator · Cloudflare Workers AI Cost Calculator
More in this group
- GPU Cloud Cost Comparison Modal vs RunPod vs Lambda monthly GPU cost + break-even.
- RunPod GPU Cost Calculator RunPod pod $/h × hours ÷ utilization vs serverless.
- fal.ai Cost Calculator fal image $/MP or $/image → per kept image & month.
- Together AI Token Cost Calculator Together serverless tokens + optional US-only.
- Fireworks Token Cost Calculator Fireworks in/cached/out + Standard/Priority/Batch.
- DeepInfra Token Cost Calculator DeepInfra tokens × Standard/Priority/Flex.
- Cloudflare Workers AI Cost Calculator Workers AI tokens + Neurons footnote.
How it works
- How much does an H100 cost on Modal?
- Modal lists H100 SXM5 at $0.001097 per second, which is $3.9492 per GPU-hour, billed only while a container runs. B200 is $0.001736/s ($6.25/h), H200 $0.001261/s, A100 80GB $0.000694/s ($2.50/h), L40S $0.000542/s, A10 $0.000306/s, L4 $0.000222/s and T4 $0.000164/s.
- Are CPU and memory billed on top of the GPU?
- Yes. CPU is $0.0000131 per physical core-second (1 core = 2 vCPU, minimum 0.125 cores) on preemptible Functions; non-preemptible and Sandboxes cost 3× ($0.00003942). Memory is $0.00000222 per GiB-second.
- What does a Stable Diffusion image cost on Modal?
- Modal's own example: about 1.5 s on an A10 with 1 core and ~4 GiB ≈ $0.000491 per image (GPU $0.000458 + CPU $0.000020 + memory $0.000013). This calculator reproduces ≈ $0.00049.
- Why include idle keep-alive seconds?
- After the last input a container stays warm for its scaledown window and that idle time is billed. Enter container starts per month × idle seconds per start (60 s default) to include it.
- How do the Starter and Team plans work?
- Starter is $0/month plus compute and includes $30 of compute; Team is $250/month plus compute and includes $100. Tick the credit box to subtract the included amount from compute.