Question 1

Is there a free plan available?

Accepted Answer

Yes. You can sign up and start using izzedo chat immediately with free models like GPT-Luna, Claude Haiku, DeepSeek or Gemini Flash — no credit card required, with about 15–25 questions in any 5-hour window. Agent mode and all 50+ connectors are included too, so you can wire up Gmail, Drive, Notion or Slack and use them with any free model. This lets you explore the interface, projects and everyday AI work before picking a paid plan for premium models, video generation and automations.

Question 2

Which plan should I pick — Hobby, Pro, Max, or Max Plus?

Accepted Answer

Match the plan to your heaviest habit, not your message count. Hobby ($6/mo) covers evenings and weekends — about 15–20 questions a day on a model mix (1–3 on top models like Opus or Fable), with room for ~45–60 in one sitting plus the odd agent task, plus video generation with MiniMax H3 Max. Pro ($12/mo) covers daily work — about 30–40 questions or 4–7 agent tasks every day, with sittings up to ~85–120 questions, and adds Kling, Wan, Gemini Omni and Grok Imagine to video. Max ($50/mo) is for AI running most of the day — about 130–180 questions or 15–30 agent tasks daily, and it unlocks every video model including Veo 3.1 and Seedance 2.0. Max Plus ($100/mo) is for routinely heavy agent and research workloads — about 250–350 questions or 30–60 agent tasks a day. Long, hours-long agent workflows belong on Max or Max Plus. Every count assumes a model mix: everyday models go several times further, and top models like Opus or Fable use the pool several times faster.

Question 3

What happens if I hit my usage limit?

Accepted Answer

You'll see a countdown until the meter frees up — the 5-hour window resets within a few hours, the weekly pool every Monday. Nothing is lost; your chats and projects stay right where they are. To keep going immediately, you can enable usage-based billing to use models past the included pool, billed only for what you use.

Question 4

Can I switch between different AI models within the same conversation?

Accepted Answer

Yes. Pick another model in the message box and send. The new model reads the whole conversation, so there's nothing to copy across. You could research with Perplexity, write with Claude Sonnet and check the numbers with Gemini Pro, all in one chat. Every model draws from the same usage pool.

Question 5

What's included in the Team and Team Max plans?

Accepted Answer

Team ($15/user/month) adds team management, role-based access, and shared projects & conversations on top of all premium models. Each member gets their own pool — about 110–150 questions, 12–25 agent tasks, or 4–6 research runs in a 5-hour window, sustainably ~40–50 questions a day on a model mix. That covers steady daily work: chat plus a handful of quick agent tasks per member. If someone runs agents continuously — hours-long workflows, many connectors, big documents — they need Team Max ($60/user/month): ~430–600 questions per window, room for 50–100 agent tasks, every video model including Veo 3.1 and Seedance 2.0, and 30 automations per member, plus early access to new features. Both plans get a discount when billed annually — 20% off Team and 17% off Team Max.

Question 6

Do I save money by paying annually?

Accepted Answer

Yes. Team is $144/user/year (effective $12/user/month, 20% off) and Team Max is $600/user/year (effective $50/user/month, 17% off). Personal plans (Hobby, Pro, Max, Max Plus) are billed monthly and can be canceled at any time.

Question 7

What do "questions", "agent tasks" and "research runs" mean in the plans?

Accepted Answer

A question is a regular chat message. The plan numbers assume a model mix: about two in three questions on everyday models like GPT-Luna, Claude Haiku or DeepSeek, the rest on GPT-Terra, Claude Sonnet or Gemini Pro. An agent task is one job the AI completes using tools and your connected apps — an inbox digest, a report pulled from your analytics, a quick site check — and it uses roughly as much as 5–10 questions, because the AI reads everything the tools return. A research run asks several models in parallel and uses a few dozen questions' worth. They all draw from one shared pool, so the plan numbers are alternatives, not separate quotas. Everyday models alone stretch the pool several times further, which makes them the best home for big agent workloads. Top models like Claude Opus or Claude Fable use it several times faster, and long chats use more than short ones, because every reply re-reads the whole conversation.

Question 8

How do the usage limits work?

Accepted Answer

Every plan has two meters, both measured by how much the AI actually reads and writes — never by counting messages. The 5-hour window covers one sitting and rolls continuously. Behind it sits the weekly pool, holding about two and a half windows' worth; it's what keeps every-day heavy use sustainable, and it resets every Monday. If you work long days daily, the weekly pool is usually the meter you'll meet first — spread across seven days it works out to roughly a third of your 5-hour numbers per day on a model mix, several times more on everyday models alone, and several times fewer on top models like Opus or Fable. Remaining usage and reset times are always visible in the app.

Question 9

Why did a few messages use so much of my limit?

Accepted Answer

Almost always because of size. Usage is measured by how much the AI reads and writes, and the whole conversation is re-read on every reply. Pasting a long document can make one message cost 10–20x a typical one, and a long agent workflow — one chat running for hours with many tool calls — can use a large share of a 5-hour window by itself. Three tips: add big documents or code to your Knowledge Base instead of pasting, start a fresh chat per sub-task (every reply gets cheaper), and run bulk work on an everyday model like GPT-Luna, Claude Haiku or DeepSeek.

Question 10

What's a context window, and why does my plan cap it?

Accepted Answer

The context window is how much of a conversation the AI can keep in mind at once — every reply re-reads the whole chat, so long conversations grow it. Each plan caps it: 50k tokens on Free, 200k on Hobby, 400k on Pro and Team, and the full model window (up to 1M tokens) on Max, Max Plus and Team Max. The app shows how full the window is in the usage indicator. When a conversation reaches the cap, start a new chat to continue — the cap is a plan feature, so it lifts when you upgrade, not with usage-based billing. For most conversations, the usage pool is the limit you'll meet first; the context cap mostly matters for marathon chats.

Start free.Every chat modelfrom $6 a month.

Free

Hobby

Pro

Max

How usage works

One pool, every model

Two meters, resetting on their own

Bigger messages use more

Long agent workflows need a big plan

If you hit a limit

Frequently asked questions

Start on the free plan. Pay when you want the premium models.