Coding plan comparison
Which coding plan is actually cheapest
Fifteen vendors, forty-odd plans, five incompatible ways of describing an allowance. Enter your cache hit rate and the page converts every plan it can into tokens per month and a blended price, and says plainly which plans cannot be converted.
- Vendors covered
- 15
- Plans compared
- 51
- Live offers
- 28
- Free models
- 75
Cheapest blended price: $0.0016 / 1M · Prices verified: 2026-10-25
Pick the view you need
Plan comparison
Every plan, converted
One row per (plan, model) pair. Pick a model family to see which vendors resell it, what each plan costs, and the blended price per million tokens that comes out.
API prices
Pay-as-you-go API prices
Pay-as-you-go list prices per 1M tokens, ranked by what 1M mixed tokens cost at your settings. Cache read and output prices decide the ranking far more than the headline input price.
Offers
Offers and discounts
Discounts, intro prices, boosted allowances and standing off-peak rates, each quoted from the vendor page it was read on. A campaign ends without the plan price changing, so it is listed here instead of in the plan table.
Free models
Models you can use for free
Every model in the snapshot that costs nothing to call, split by how you get it: a hosted API, your own hardware, or a paid plan that includes it. Expiry dates are the vendor's, not ours.
Offers running now
Discounts, intro prices, boosted allowances and standing off-peak rates, each quoted from the vendor page it was read on. A campaign ends without the plan price changing, so it is listed here instead of in the plan table.
OpenCode
- DeepSeek V4.1 Flash is priced per time of day: the off-peak row is half the peak rate, and both are listed.
- DeepSeek V4 Pro is priced per time of day: the off-peak row is half the peak rate, and both are listed.
- Gemini 4 Argon: 50% off list, already applied in the price shown.
Command Code
- deepseek-v4.1-flash: boosted allowance of $10 on Go while capacity lasts.
- glm-5.3-flash: boosted allowance of $10 on Go while capacity lasts.
- ling-3.1-flash: boosted allowance of $100 on GOAT while capacity lasts.
- minimax-m3: 2× usage on Go while capacity lasts.
OpenDesign
- A week-long campaign bills DeepSeek V4.1 Flash at off-peak rates.
- First month $8 instead of $10.
火山方舟 Volcengine
- kimi-k2.8-preview: limited-time 40% off, so the same allowance buys more tokens.
- kimi-k3: limited-time 50% off, so the same allowance buys more tokens.
- kimi-k3.1-pro: limited-time 50% off, so the same allowance buys more tokens.
- deepseek-v4.1-flash: limited-time 50% off, so the same allowance buys more tokens.
- deepseek-v4.1-flash-economy: limited-time 50% off, so the same allowance buys more tokens.
- First 2 months ¥9.9 instead of ¥40.
阿里云百炼 Alibaba Model Studio
- First month ¥39.9 instead of ¥200.
Z.ai
- Off-peak usage is charged at 50% of the standard credit rate.
Cerebras Code
- A limited free trial runs GLM-4.7 with capped tokens and requests.
GitHub Copilot
- Copilot Free and Student include a small AI-credit allowance, in-product only.
百度千帆 Qianfan
- Off-peak calls (22:00-08:00) on 3 named models cost 80% less until 2026-11-30.
月之暗面 Kimi
- Annual billing is about 20% cheaper: Plus ¥79/month instead of ¥99.
MiniMax
- First month $11 instead of $22.
Anthropic Claude
- Claude Free is a $0 tier for chat on web, desktop and mobile, without Codex-style usage.
- Paying yearly brings Pro to $17/month, about 15% below the monthly price.
- Universities can buy discounted access for every member; the percentage is not published.
OpenAI ChatGPT
- ChatGPT Free includes Codex on quick tasks, with GPT-6 Luna at Standard speed.
- Business seats are $20/user billed yearly versus $25 monthly, about 20% less.
Cursor
- Cursor Free is a $0 tier with limited agent and chat requests, in-product only.
Read the numbers like this
- A plan is not a price per token. The same $20 buys 1.5B tokens of a cheap open model or 40M tokens of a frontier model, and several plans grant a different allowance per model.
- Request-based plans are converted with a stated tokens-per-request assumption, because a request count is not a token count. Treat those rows as an estimate and check your own task size.
- Credits are converted with the vendor's own formula (credits = weighted tokens / divisor, peak/off-peak multipliers applied), so the number is exact for that model but moves with time-of-day.
- Claude, ChatGPT, Cursor, Kimi and MiniMax publish no convertible allowance at all. They stay in the table, sorted last, with their price — an honest gap beats an invented number.
FAQ
Why do some rows show a token count and others say not quantified?
Because plans do not sell tokens. Some sell a dollar allowance (OpenCode, Command Code, OpenDesign), some a token volume (Synthetic, Cerebras, Baidu), some vendor credits with per-model multipliers (Z.ai, Volcengine Agent Plan), some a request count (Volcengine, Alibaba), and the biggest names sell an unquantified "usage" (Claude, ChatGPT, Cursor, Kimi, MiniMax). The Allowance column says which, so an inferred number is never mistaken for a published one.
Why not just compare the price per million tokens?
Input, cached read, cached write and output are priced separately, and their ratios differ by an order of magnitude between providers. A single blended figure at stated assumptions is the only number that compares plans honestly.
I only care about one model. What do I do?
Pick the model family first (DeepSeek, GLM, Claude, Qwen...). The page then lists every vendor that resells that family, what each plan costs, how many tokens the plan buys at your cache hit rate, and the resulting blended unit price.
How accurate is the token estimate?
Request counts are the one estimate on the page: they are multiplied by a stated tokens-per-request figure that you can adjust. Every other conversion uses a number the vendor publishes, and every row links to its source page so you can check it.
Where does this data come from?
Unit prices come from LiteLLM and OpenRouter; subscription prices and allowances are parsed from each vendor's own plan page (or declared with an anchor guard when the page only renders prices in the browser). The verified date is on every row — prices move, so re-check before committing to an annual plan.