Is Claude Code free? What it costs, and when to pay for it anyway

What Claude Code costs

Everything in this section comes from Anthropic's pricing page and Claude Code documentation, read on 18 September 2026.

The same page says where Claude Code's usage comes from:

"Claude Code is included in all paid plans. It shares the same usage limits as the rest of your plan, so your work in the terminal and your chats draw from one pool."

On the size of that pool, Anthropic's FAQ says usage varies with what you ask and which model answers, "so there's no fixed message count." Pro is described as at least 5x the usage of Free per five-hour session, and Max as 5x or 20x the usage of Pro. Those are ratios with nothing at the bottom of them, so nobody outside Anthropic can turn a plan into a number of finished tasks. We do not try.

The one hard number comes from the pay-as-you-go side. Anthropic's cost documentation reports that "Across enterprise deployments, the average cost is around $13 per developer per active day and $150-250 per developer per month, with costs remaining below $30 per active day for 90% of users." That is what Claude Code costs when you pay for the tokens directly.

What Clixad does differently

Clixad does the same job in the terminal, editing files and running commands, and it asks before either unless you switch that off. The difference is the bill.

Tokens are the unit. Every model call is metered from its real token count at a published rate, 100,000 credits to one US dollar of list value. A small job costs a small amount, and the price of each model is public before you pick it.

Five cheap models are free every day. GPT-5 Nano, Gemini 2.5 Flash-Lite, Claude Haiku 5.5, DeepSeek V4 Flash and Gemini 3.5 Flash-Lite cost nothing up to 60,000 credits of usage per account per day. That allowance sits inside a daily budget shared by everybody, so on a busy day the free turns can run out sooner. Free usage adds and deducts nothing. Past either limit those models cost credits again until the next day.

The rest of the catalogue runs on credits. Thirteen models from five vendors sit on one balance, Anthropic's among them, and /model switches between them mid-session. A new account starts with 25,000 credits. After that you earn credits by completing an advertiser action on an offerwall page in a browser tab, or you buy them; the smallest pack is $3 for 280,000 credits.

Earning is uneven. Credits arrive on completion. The most common outcome of starting a survey is a screenout: a few questions in, the survey decides you are not the respondent it wants and ends with nothing credited, or with a small bonus. From our own ledger, completed offers on our CPX wall have credited from 9,000 to 114,000 credits on completion, with a median of 51,000: 26 completions recorded in our ledger between 27 July 2026 and 12 September 2026, as of 14 September 2026. The ledger does not record how long an offer took.

The session shows ads. While a model call runs, one line above the input box can show a sponsored message marked Ad, and up to two more places on screen can show one marked the same way. Nothing you type takes part in choosing them, and CLIXAD_SPONSOR=0 stops every request for them being made at all. Claude Code shows no advertising, and that counts in its favour.

Your project files mostly carry over. Clixad reads project instructions from AGENTS.md and from its own CLIXAD.md. It does not read CLAUDE.md, so copy that file across if you switch. MCP servers declared in a project's .mcp.json use the mcpServers object Claude Code writes, and Clixad reads that file as it is.

One agent task, priced

A plan price leaves out what you get, so this blog prices one fixed workload: a single instruction that an agent turns into twelve model calls, starting at 12,000 input tokens and resending the whole conversation every round. The shape comes from the tier-limits post.

Read the Claude rows first. One of them is free: Claude Haiku 5.5 is in the daily allowance, and 60,000 credits cover 10 whole tasks on it, as many as on Gemini 2.5 Flash-Lite. The larger Claude models are not. The cheapest of them, Sonnet 5.5, costs 115,380 credits a task, more than twice what one median completed offer at 51,000 credits pays, and it is a premium model, which the 25,000-credit signup bonus does not pay for. A full day at the earning cap buys 4 tasks on Sonnet 5.5 and 2 on Opus 5.5. Fable 5.1 costs more per task than the whole cap, so a day of earning does not buy one.

The top three rows are where the free part lives: 19 whole tasks a day on GPT-5 Nano, 10 on Gemini 2.5 Flash-Lite and 10 on Claude Haiku 5.5, with no earning at all. Whether a cheap model is good enough depends on your code, and trying it costs nothing.

Buying credits changes where the balance comes from and nothing else. The Starter pack, $5 for 500,000 credits, buys exactly what a full day at the earning cap does: 4 tasks on Claude Sonnet 5.5.

Who Clixad fits

Someone who codes in bursts. Credits do not expire, so a quiet week costs nothing and the balance is still there the week after. A subscription bills the quiet week at the price of the busy one.

Work that cheap models can carry. The daily allowance handles small tasks for free. When a problem needs a frontier model you pay for that one task, 230,760 credits on Claude Opus 5.5, with no monthly commitment behind it.

Someone who wants several vendors on one balance. Claude, GPT, Gemini, DeepSeek and Kimi models sit behind the same /model picker, priced per token. Trying a second model on a problem the first one fumbled costs that call and nothing more.

Someone without a card to put on file. The account is a GitHub login, and neither the free allowance nor the offerwall asks for payment details.

Who should pay for Claude Code

Anyone who codes all day on Claude models. Anthropic builds its top plan for exactly that reader, in its own words: "The Max plan is built for people who work with Claude throughout the day and want longer, uninterrupted sessions."

Our own numbers point the same way. Anthropic's average for enterprise developers paying by the token is around $13 per active day. Our whole daily earning cap is 500,000 credits, which is $5 of list value, less than half of that average day, and reaching it takes several completed offers. Buying credits does not close the gap cheaply either: bought credits pay for tokens at provider cost with a 1.5 markup, which the funding post sets out. Anthropic's own per-token average of $150-250 per developer per month already sits above the $100 at which Max starts.

Claude Code also does more than we do. Its documentation covers hooks, skills, plugins and agent teams, and Clixad has none of them. If those are part of how you work, stay with Claude Code.

Other free agents exist, and they are built differently from ours. FreeBuff grants a daily allowance of hours on its own model picker, and the shortlist of agents that include model access covers the rest. If your work leans on OpenAI's models, the Codex comparison runs the same numbers against ChatGPT's plans.

FAQ