Comparisons with published methodology Every claim dated & sourced

Local LLM stacks · Tool review

Ollama pricing, free local models plus a $20 cloud tier

Updated 2026-09-28 · markdown mirror

Verdict

Ollama costs nothing to run models on your own hardware, and that holds on every plan including no plan at all. The money is entirely in the cloud service, Pro at $20 a month with $60 of usage credits, Max at $100 with $300, and Team at $500 with $1,000 shared across unlimited users. If your models fit on your machine, every paid tier is optional; if you need frontier-scale models without buying hardware, Pro is the entry point.

Local LLM stacks: the compared products as a logo collection grid

Visit Ollama

Ollama pricing (every row carries its source)
TierPriceVerifiedSource
Free $0/mo Documented at the source Checked 2026-09-28 ollama.com /cloud Local models always free, cloud use is pay-as-you-go with starter credits 2026-09-28 ollama.com
Pro $20/mo Documented at the source Checked 2026-09-28 ollama.com /cloud $16.67 a month billed annually ($200 a year), includes $60 of monthly usage credits 2026-09-28 ollama.com
Max $100/mo Documented at the source Checked 2026-09-28 ollama.com /cloud Includes $300 of monthly usage credits and 10 concurrent requests 2026-09-28 ollama.com
Team $500/mo Documented at the source Checked 2026-09-28 ollama.com /cloud Unlimited users with $1,000 of shared monthly usage credits 2026-09-28 ollama.com

● tested by us · ● documented at the source · ● unconfirmed, queued in /confirm.txt — hover or tap a dot for the source and date

What it is

Ollama is an open-source tool for running open models locally with one command and an OpenAI-compatible API. This page covers its pricing only; the local-versus-desktop comparison with LM Studio lives in our Ollama vs LM Studio review.

The price ladder, with numbers

The whole ladder, verified 28 September 2026 [1]:

Plan Price Included usage Concurrent requests
Free $0 Starter credits 1
Pro $20/mo, or $200/yr $60/mo 3
Max $100/mo $300/mo 10
Team $500/mo $1,000/mo shared 10
Enterprise Custom Custom Custom

Annual billing exists only on Pro at $200 a year, which the pricing page presents as about $16.67 a month. Max and Team are monthly only at our check, and Team carries an early-access flag.

Local versus cloud, where the money actually is

The vendor’s own framing is “local models are always free” [1], and the billing FAQ confirms running models on your own hardware is unlimited on every plan [1]. Every paid dollar buys cloud capacity, hosted model access, and concurrency, not software features. The practical test is where your workload fits: an 8B model on a decent laptop needs no subscription, while frontier-scale models beyond your VRAM are what Pro and Max monetise.

Credits, resets, and off-peak rates

Plans include a dollar amount of usage each month, measured in tokens at per-model rates, and draw included credits before touching any purchased balance [1]. Included usage does not roll over, it refreshes at each monthly reset, and Ollama emails you at 90 percent of included usage on paid plans. Cloud token prices fall off-peak, outside 12:00 to 18:00 UTC on weekdays and all day on weekends, which halves the example deepseek-v4.1-flash rate to $0.15 per million input tokens [1]. Batch overnight jobs off-peak and the included credits stretch further.

Don’t pay for it if

Your models run locally, because no plan feature changes that. If you do go cloud, match the tier to steady monthly usage rather than peak curiosity, since credits expire monthly and the concurrency jump that matters for multi-agent setups sits at Max, ten requests against Pro’s three [1].

FAQ

How much does Ollama cost?

The software is free and open source, and running models on your own hardware is always free and unlimited on every plan. The cloud plans are Free at $0 with starter usage credits, Pro at $20 a month ($200 billed annually, about $16.67 a month) with $60 of monthly usage credits, Max at $100 a month with $300 of credits, and Team at $500 a month with $1,000 of credits shared across unlimited users. Enterprise is custom. Verified on the vendor pricing page, 28 September 2026 [1].

Is Ollama free to use?

Yes for local models, which run free and unlimited on your own hardware with no account needed. The cloud service is pay as you go, the Free plan includes starter credits for a smaller set of starter models, and buying usage credits unlocks all cloud models. Verified 28 September 2026 [1].

What does Ollama Pro include?

Pro costs $20 a month, or $200 a year billed annually, and includes $60 of usage credits each month, access to larger pro models, three concurrent requests, and multiple models running concurrently. The included usage resets monthly on the day your subscription started. Verified 28 September 2026 [1].

Do Ollama usage credits roll over?

No. Included monthly usage on Pro, Max, and Team refreshes at each monthly reset and unused credits do not roll over, the vendor states this directly in its billing FAQ [1]. Plans draw included credits first, then fall back to any purchased credit balance. Verified 28 September 2026.

What is Ollama off-peak pricing?

Discounted per-token rates on cloud models outside weekday peak hours, which are 12:00 to 18:00 UTC, with off-peak applying all day on weekends. For example deepseek-v4.1-flash drops from $0.30 to $0.15 per million input tokens and from $1.20 to $0.60 per million output tokens off-peak. Verified 28 September 2026 [1].

Sources

  1. Ollama pricing — Vendor pricing page, Free 0 dollars with starter credits, Pro 20 dollars a month or 200 annually with 60 dollars of monthly credits, Max 100 dollars with 300, Team 500 dollars with 1000 shared, model token table with off-peak rates, concurrency 1, 3, 10, 10, no rollover stated in billing FAQ, verified 2026-09-28
  2. Ollama homepage — Vendor homepage stating local models are always free and usage resets monthly, verified 2026-09-28

Numbered markers in the text refer to this list. External sources are cited, not affiliated.