> Modal
Modal is serverless compute for AI workloads rather than a token-billed model API: you pay per second for CPU, memory and GPU time (H100 SXM5 $0.001097/sec, B200 $0.001736/sec, CPU $0.0000131 per core per second) and never for idle resources. The Starter plan is $0 plus compute and includes $30 per month of free compute with 3 seats, 100 containers and 10 GPU concurrency; Team is $250 per month with $100 per month of free compute. Volumes include 1 TiB per month free, then $0.09/GiB per month.
// pros
- Starter includes $30 per month of free compute at no plan fee, with 3 workspace seats, 100 containers, 10 GPU concurrency, region selection and unlimited secrets and custom images.
- Billing is per second with no charge for idle resources: CPU at $0.0000131 per physical core per second (0.125-core minimum per container), memory at $0.00000222 per GiB per second, H100 SXM5 at $0.001097 per second.
- Volumes include 1 TiB per month free before the $0.09 per GiB per month rate applies.
- Academic researchers (graduate students, labs) can apply for up to $10,000 in free compute credits, and early-stage startups can apply for separate credit grants.
- Environment-level budgets are available on Starter, and Team and Enterprise add programmatic billing reports through the modal.billing API and CLI.
- Cost multipliers are opt-in and published rather than default: region selection adds 1.15-1.75x base prices and non-preemptible execution costs 3x base prices.
// cons
- $30 per month of free compute is a small allowance against GPU rates: at the published B200 rate of $0.001736 per second it is about 4.8 GPU-hours per month, so the credit suits testing rather than serving.
- A payment method is required to use Modal at all, per the billing docs, so the Starter grant is not card-free.
- Starter limits constrain production use: 1-day log retention, 5 deployed cron jobs, 3 seats, 200 deployed apps and 3 deployment rollback versions.
- Included token usage is 0 on every plan, so any LLM token spend sits on top of the included compute.
- Incremental usage charges are auto-charged mid-cycle once set thresholds are crossed, so spend does not move strictly on the monthly cycle.
Use Modal when your workload is a container or GPU job you can define in Python and per-second idle-free billing matters more than an off-the-shelf chat endpoint; the $30 per month Starter credit covers evaluation and small services, and it pairs well in a router as the execution layer for models other providers bill per token to serve.
written by Mike Fleming, Uprouter editorial · verified against official sources September 16, 2026
How we estimated this: the free allowance is converted to a dollar value using the provider’s own pay-as-you-go rates for the models a typical developer would reach for. When the allowance is metered in requests, tokens or minutes rather than dollars, we assume median usage of the cheapest capable model. Because assumptions are involved, the figure carries the est. flag — treat it as an order of magnitude, not a quote.
note: Starter plan includes $30 per month of free compute, stated on the Modal pricing page (Team includes $100 per month on the $250 plan); academics can apply separately for up to $10,000 in credits.
| Plan | Type | Monthly | Input /1M | Output /1M | Note |
|---|---|---|---|---|---|
| Free tier (source-reported) | free | — | — | — | ~$5/month credits on sign up; ~$30/month after adding a payment method. A payment method on file is required to use Modal; billed by compute time. |
| Free tier | free | — | Free | Free | $30/mo compute, 10 GPU concurrency |
| Pay-as-you-go | payg | — | $1.10 | — | per-second billing; model costs self-managed |
| Starter (Free) | free | — | — | — | $30/month in compute credits; 10 GPU concurrency. |
| H100 SXM PAYG | payg | — | — | — | Estimated hourly rate based on serverless usage ($0.044/min). |
Pricing normalized from public sources — always verify with the provider.
// model prices
| Model | Input /1M | Output /1M | Context |
|---|---|---|---|
| No fixed free-model list (self-deploy any supported model, billed by compute) | — | — | — |
$ Does Modal have a free tier?
Yes — Modal offers a free tier that we value at roughly $30.00 in free credits (estimate — derived from rate-limited usage at pay-as-you-go prices). Always confirm current limits on Modal's official pricing page — free tiers change without notice.
$ How much does Modal cost per 1M tokens?
Modal does not publish simple per-1M-token pricing (it may bill per request, per output, or via a subscription plan). Check their official pricing page for exact figures.
$ Is Modal safe to use?
Uprouter rates Modal as low risk. Low to medium. The free grant is explicit and recurring ($30 per month on Starter, $100 per month included on the $250 Team plan) and billing is documented per second, with the two cost multipliers - region selection and non-preemptible execution - opt-in and published, which keeps spend predictable. Constraints are structural rather than policy-based: a payment method is mandatory, Starter log retention is one day, cron jobs are capped at five, and crossing incremental-usage thresholds triggers mid-cycle charges. Risk ratings are editorial, evidence-based, and never influenced by affiliate relationships.
Raw community sentiment, shown unblended. Votes never feed into objective data fields like price, risk, or status — and never into sort order.
No published reviews yet — be the first.
// more_informationprice history · live status · risk & confidence · code snippets
price_history
No recorded changes yet.
live_status
current: upNo probes recorded yet.
Uprouter runs its own probes; best-effort, not a guarantee. Probe results depend on our network location and cadence — for production, always use the provider’s own status page.
risk_&_confidence
Low to medium. The free grant is explicit and recurring ($30 per month on Starter, $100 per month included on the $250 Team plan) and billing is documented per second, with the two cost multipliers - region selection and non-preemptible execution - opt-in and published, which keeps spend predictable. Constraints are structural rather than policy-based: a payment method is mandatory, Starter log retention is one day, cron jobs are capped at five, and crossing incremental-usage thresholds triggers mid-cycle charges.
Confidence reflects how much verifiable evidence (official docs, probe history, community reports) backs this entry. Risk is our editorial assessment of operator reliability and terms — not financial advice.
code_snippets
# Modal's own API — get a key from their dashboard
curl https://YOUR-PROVIDER-ENDPOINT/chat/completions \
-H "Authorization: Bearer THEIR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "no-fixed-free-model-list-self-deploy-any-supported-model-billed-by-compute",
"messages": [{ "role": "user", "content": "Hello" }]
}'Direct to the provider — Uprouter receives nothing; verify their current pricing before purchasing.
Ranked by neutral similarity signals only — live status, free-tier shape and free-credit proximity. Commercial relationships never influence this list.