Skip to content
tokenblackhole.

verboo.ai: the tokens really are unlimited. Two concurrent requests is the actual plan.

@aimaster36582 min read4.0 / 5

Site reviews

Screenshot of the verboo.ai home page
verboo.ai home page, checked 2026-09-20

Verdict

$69 a month for unlimited GLM 5.3 Flash, DeepSeek V4.1 Flash and DeepSeek V4 Flash. Every tier from $24 to $269 carries the same two concurrent requests, so paying more buys a bigger model list, not more parallelism. If you fan out subagents, money does not raise that ceiling.

Unlimited tokens is not marketing here. It is the actual product, and it works. What is not unlimited is how many things you can run at once, and that is the number to decide on.

What you get

The $69 Pro tier runs GLM 5.3 Flash, DeepSeek V4.1 Flash and DeepSeek V4 Flash with no token budget at all. Measured at check time: DeepSeek V4 Flash 89.8 tok/s, DeepSeek V4.1 Flash 82.6 tok/s, GLM 5.3 Flash 36.8 tok/s. 90 tok/s is genuinely usable.

Unlimited is not unconstrained. What actually binds is two concurrent requests and 40 requests a minute, on every single tier.

The catch is the ladder, not the price

$24, $69, $129 and $269 all carry the same 40 requests a minute and the same two concurrent requests. More money buys a larger model set and nothing else. $24 gets Qwen3.8 27B; $69 adds GLM 5.3 Flash and DeepSeek V4.1 and V4 Flash; $129 adds DeepSeek V4 Pro; $269 adds GLM 5.3.

  • An agent that fans out subagents queues here. $269 still gets you two concurrent requests.
  • Capacity is capped per tier and it does sell out. Pro showed 843 of 1000 spots left at check time.
  • The same site also sells a prepaid per-token API from $0.12 per million input tokens. Unlimited is the subscription, not the API.

Verdict

If you burn a lot of tokens and only ever run two things at once, nothing at this price comes close. If you run wide, paying more here fixes nothing — parallelism is a thing you have to buy somewhere else.

The word unlimited is attached to the tokens. It is not attached to the concurrency.

Questions

Are the tokens actually unlimited?
Yes. There was no token budget at check time. What you get instead is two concurrent requests and 40 requests a minute, on every tier.
Does a higher tier let me run more at once?
No. $24 and $269 carry the same two concurrent requests. The tier changes which models you can reach and nothing else.
How fast is GLM 5.3 Flash here?
36.8 tok/s at check time. In the same session DeepSeek V4 Flash measured 89.8 tok/s and DeepSeek V4.1 Flash 82.6 tok/s.

The terms, as verified

Unlimited tokens

Read from the vendor's own pricing page. The review above is a judgement about these numbers; it does not restate them, so the two cannot drift apart.

$69.00

$69/month · Pro

Unlimited tokens

What actually limits you: 2 concurrent requests, 40 req/min

843 of 1000 spots left when checked

Models GLM 5.3 Flash, DeepSeek V4.1 Flash, DeepSeek V4 Flash

Worth knowing

  • Unlimited does not mean unconstrained. The binding limit is two concurrent requests on every tier — an agent that fans out subagents will queue, and no amount of money raises that ceiling.
  • The ladder changes which models you get, not how fast you can call them. $24 buys Qwen3.8 27B; $69 adds GLM 5.3 Flash and DeepSeek V4.1 and V4 Flash; $129 adds DeepSeek V4 Pro; $269 adds GLM 5.3.
  • Capacity is capped per tier and it does sell out — Pro showed 843 of 1000 spots left at check time.
  • The same site sells a prepaid per-token API from $0.12 per million input tokens. Unlimited is the subscription, not the API.
  • Runs on their own GPUs with zero data retention, on Brazilian servers. Monthly or yearly.

Source: verboo.aiChecked 2026-09-20

More site reviews