All plans
44 paid plans, 38 with a usage figure. Prices are US monthly before tax, per seat for team plans. Usage is the plan’s largest included allowance at API list prices. Value per dollar is coding turns of balanced work (5% frontier, 15% hard, 60% normal, 20% light) per dollar, each kind on the model that stretches the allowance furthest; the models used are listed under it. Limits were read from the linked provider pages on Nov 12, 2026.
| Plan | Price / mo | Largest allowance on | Top score | Usage / mo | Per dollar | Value / $ | Usage data | Limit and sources |
|---|---|---|---|---|---|---|---|---|
Claude ProClaude Code | $20 | Opus 5.5 | 59.1 | $669–$1.4k | 33–69× | 38–78Opus 5.5 max · Opus 5.5 xhigh · Sonnet 5.5 xhigh · Haiku 5.5 high | Measured | More usage than Free, per five hours and per week. Fable is not included; it draws on paid usage credits.Anthropic publishes no absolute limit. Range from three independent measurements of real usage valued at API list prices with heavy caching: wokey.ai (meter readings across subscriber accounts, November 2026), SemiAnalysis (controlled calls against the usage meters, October 2026, quoted from secondary coverage of its paywalled charts) and Real API Pricing (Opus 5.5 usage read against the weekly meter, September to November 2026). The monthly API credits on Max and Ultra are not counted: they cannot be spent in Claude Code. Claude pricing · Fable models on your plan · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (SemiAnalysis) · Real API Pricing (measured allowances) |
Claude Pro yearlyClaude Code | $16.67 | Opus 5.5 | 59.1 | $669–$1.4k | 40–83× | 45–93Opus 5.5 max · Opus 5.5 xhigh · Sonnet 5.5 xhigh · Haiku 5.5 high | Measured | Claude Pro billed yearly: $200 up front. Same limits as Pro monthly.Anthropic publishes no absolute limit. Range from three independent measurements of real usage valued at API list prices with heavy caching: wokey.ai (meter readings across subscriber accounts, November 2026), SemiAnalysis (controlled calls against the usage meters, October 2026, quoted from secondary coverage of its paywalled charts) and Real API Pricing (Opus 5.5 usage read against the weekly meter, September to November 2026). The monthly API credits on Max and Ultra are not counted: they cannot be spent in Claude Code. Claude pricing · Fable models on your plan · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (SemiAnalysis) · Real API Pricing (measured allowances) |
Claude Team StandardClaude Code | $20 | Opus 5.5 | 59.1 | $836–$1.7k | 42–87× | 47–97Opus 5.5 max · Opus 5.5 xhigh · Sonnet 5.5 xhigh · Haiku 5.5 high | Derived | Per seat, billed annually ($25 monthly), for teams of 2 to 150. 1.25× Pro usage per session.Anthropic publishes no absolute limit. Range from three independent measurements of real usage valued at API list prices with heavy caching: wokey.ai (meter readings across subscriber accounts, November 2026), SemiAnalysis (controlled calls against the usage meters, October 2026, quoted from secondary coverage of its paywalled charts) and Real API Pricing (Opus 5.5 usage read against the weekly meter, September to November 2026). The monthly API credits on Max and Ultra are not counted: they cannot be spent in Claude Code. Team seats are scaled from the measured Pro range by the multiples in Anthropic's support article. One Anthropic help article lists Premium at $150 a seat; the pricing page's $100 annual price is used. Claude pricing · Fable models on your plan · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (SemiAnalysis) · Real API Pricing (measured allowances) · Team premium seats |
Claude Team PremiumClaude Code | $100 | Opus 5.5 | 64.6 | $4.2k–$8.7k | 42–87× | 59–109Fable 5.2 max · Opus 5.5 xhigh · Sonnet 5.5 xhigh · Haiku 5.5 high | Derived | Per seat, billed annually ($125 monthly), for teams of 2 to 150. 6.25× Pro usage per session; Fable for up to half of the weekly limit.Anthropic publishes no absolute limit. Range from three independent measurements of real usage valued at API list prices with heavy caching: wokey.ai (meter readings across subscriber accounts, November 2026), SemiAnalysis (controlled calls against the usage meters, October 2026, quoted from secondary coverage of its paywalled charts) and Real API Pricing (Opus 5.5 usage read against the weekly meter, September to November 2026). The monthly API credits on Max and Ultra are not counted: they cannot be spent in Claude Code. Team seats are scaled from the measured Pro range by the multiples in Anthropic's support article. One Anthropic help article lists Premium at $150 a seat; the pricing page's $100 annual price is used. Claude pricing · Fable models on your plan · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (SemiAnalysis) · Real API Pricing (measured allowances) · Team premium seats |
Claude MaxClaude Code | $100 | Opus 5.5 | 64.6 | $3.3k–$8.3k | 33–83× | 47–100Fable 5.2 max · Opus 5.5 xhigh · Sonnet 5.5 xhigh · Haiku 5.5 high | Measured | 5× Pro usage per five hours and per week. Fable is included for up to half of the weekly limit.Anthropic publishes no absolute limit. Range from three independent measurements of real usage valued at API list prices with heavy caching: wokey.ai (meter readings across subscriber accounts, November 2026), SemiAnalysis (controlled calls against the usage meters, October 2026, quoted from secondary coverage of its paywalled charts) and Real API Pricing (Opus 5.5 usage read against the weekly meter, September to November 2026). The monthly API credits on Max and Ultra are not counted: they cannot be spent in Claude Code. Max has one direct measurement (SemiAnalysis); the low end is five times the lower Pro figure. Fable: the measured Fable 5.1 figure up to half of the plan's low end. Claude pricing · Fable models on your plan · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (SemiAnalysis) · Real API Pricing (measured allowances) |
Claude UltraClaude Code | $200 | Opus 5.5 | 64.6 | $7k–$14k | 35–72× | 49–91Fable 5.2 max · Opus 5.5 xhigh · Sonnet 5.5 xhigh · Haiku 5.5 high | Measured | 10× Pro usage per week, no five-hour limit for verified accounts. Fable is included for up to half of the weekly limit.Anthropic publishes no absolute limit. Range from three independent measurements of real usage valued at API list prices with heavy caching: wokey.ai (meter readings across subscriber accounts, November 2026), SemiAnalysis (controlled calls against the usage meters, October 2026, quoted from secondary coverage of its paywalled charts) and Real API Pricing (Opus 5.5 usage read against the weekly meter, September to November 2026). The monthly API credits on Max and Ultra are not counted: they cannot be spent in Claude Code. Fable: the measured Fable 5.1 figure up to half of the plan's low end. Claude pricing · Fable models on your plan · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (SemiAnalysis) · Real API Pricing (measured allowances) |
ChatGPT GoCodex | $8 | GPT-5.6 Terra | 32.3 | — | — | — | Not published | Codex access is described as “Limited”; no message or credit allowance is published.No absolute allowance is published, so this plan has a price and a model list but no usage figure. Codex pricing and limits · ChatGPT pricing · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (finopsllm) |
ChatGPT PlusCodex | $20 | GPT-6 Astra | 56.9 | $160–$300 | 8–15× | 7.6–14GPT-6.1 Sol max · GPT-6.1 Terra xhigh · GPT-6 Luna maxlittle on frontier work | Measured | Local Codex messages per five hours: 5–45 on GPT-6 Astra, 15–160 on GPT-6.1 Sol, 35–285 on GPT-6.1 Terra, 350–3,000 on GPT-6 Luna. Weekly limits may apply.Range from three independent measurements of real usage valued at API list prices: wokey.ai ($46 a week, November 2026), finopsllm (median $67 a week, October 2026) and SemiAnalysis ($162–211 a month, October 2026). A derivation from OpenAI's published five-hour message limits gives the same range. OpenAI moves to a shared token allowance in December 2026, after which this needs re-measuring. Codex pricing and limits · ChatGPT pricing · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (finopsllm) |
ChatGPT Pro 100Codex | $100 | GPT-6 Astra | 56.9 | $1.6k–$3k | 16–30× | 15–28GPT-6.1 Sol max · GPT-6.1 Terra xhigh · GPT-6 Luna maxlittle on frontier work | Derived | No five-hour limit. 10× Plus, per OpenAI's Codex lead on 8 November 2026; no figure is on OpenAI's own pages.The measured Plus range times 10. Allowances were doubled on 29 October 2026 and no measurement exists since. Codex pricing and limits · ChatGPT pricing · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (finopsllm) · Pro multiples, 8 Nov 2026 (OpenAI Codex lead, via secondary report) |
ChatGPT Pro 200Codex | $200 | GPT-6 Astra | 56.9 | $2.5k–$6k | 13–30× | 12–28GPT-6.1 Sol max · GPT-6.1 Terra xhigh · GPT-6 Luna maxlittle on frontier work | Derived | No five-hour limit. 20× Plus, per OpenAI's Codex lead on 8 November 2026.Low end: wokey.ai's measurement of $581 a week (about $2,519 a month) on 7 accounts, in a week that straddles the 29 October increase. High end: the measured Plus range times 20. Codex pricing and limits · ChatGPT pricing · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (finopsllm) · Pro multiples, 8 Nov 2026 (OpenAI Codex lead, via secondary report) |
ChatGPT Pro 500Codex | $500 | GPT-6 Astra | 56.9 | $8k–$15k | 16–30× | 15–28GPT-6.1 Sol max · GPT-6.1 Terra xhigh · GPT-6 Luna maxlittle on frontier work | Derived | No five-hour limit. 50× Plus, per OpenAI's Codex lead on 8 November 2026.The measured Plus range times 50. No measurement exists since allowances were doubled on 29 October 2026. Codex pricing and limits · ChatGPT pricing · Measured limits, Nov 2026 (wokey.ai) · Measured limits, Oct 2026 (finopsllm) · Pro multiples, 8 Nov 2026 (OpenAI Codex lead, via secondary report) |
Google AI ProAntigravity | $19.99 | Gemini 3.8 Flash | 59.1 | $517–$536 | 26–27× | 8.5–25Gemini 4 Hydrogen xhigh · Gemini 4 Hydrogen high · Gemini 3.8 Flash mediumlittle on frontier work | Measured | A quota that refreshes every five hours until a weekly limit, plus a separate, small weekly Argon quota. No size is published.Measured: Pro's shared Gemini pool read against logged tokens priced at Gemini list rates, twice (Real API Pricing, September 2026), $119–123 a week. Gemini 3.8 Flash draws on it at API prices. Hydrogen is a day old and its rate is unknown; its range runs from a third of the pool to all of it. Claude models in Antigravity have their own unpublished limit and are not valued. Antigravity plans · Changes to Antigravity plans (5x / 20x) · Antigravity models · Gemini 4 Hydrogen in Antigravity · Real API Pricing (measured allowances) |
Google AI Ultra 5×Antigravity | $99.99 | Gemini 3.8 Flash | 60.0 | $2.6k–$2.7k | 26–27× | 15–52Gemini 4 Carbon high · Gemini 4 Carbon medium · Gemini 4 Hydrogen high · Gemini 3.8 Flash medium | Anecdotal | 5× the tokens of Pro, refreshed every five hours until a weekly limit.The measured Pro pool (Real API Pricing, $119–123 a week) times Google's official 5×. Gemini 3.8 Flash draws on it at API prices. Carbon and Argon use it up about three times as fast as their API price implies, according to one secondhand user report, so they are valued at a third of the pool, shown from half to double that. Hydrogen is a day old and its rate is unknown; its range runs from Argon’s rate to Flash’s. Claude models in Antigravity have their own unpublished limit and are not valued. Antigravity plans · Changes to Antigravity plans (5x / 20x) · Antigravity models · Gemini 4 Hydrogen in Antigravity · Real API Pricing (measured allowances) |
Google AI Ultra 20×Antigravity | $199.99 | Gemini 3.8 Flash | 60.0 | $10k–$11k | 52–54× | 31–103Gemini 4 Carbon high · Gemini 4 Carbon medium · Gemini 4 Hydrogen high · Gemini 3.8 Flash medium | Anecdotal | 20× the tokens of Pro, refreshed every five hours until a weekly limit.The measured Pro pool (Real API Pricing, $119–123 a week) times Google's official 20×. Gemini 3.8 Flash draws on it at API prices. Carbon and Argon use it up about three times as fast as their API price implies, according to one secondhand user report, so they are valued at a third of the pool, shown from half to double that. Hydrogen is a day old and its rate is unknown; its range runs from Argon’s rate to Flash’s. Claude models in Antigravity have their own unpublished limit and are not valued. Antigravity plans · Changes to Antigravity plans (5x / 20x) · Antigravity models · Gemini 4 Hydrogen in Antigravity · Real API Pricing (measured allowances) |
Copilot ProGitHub Copilot | $10 | Fable 5.2 | 64.6 | $15 | 1.5× | 3.3Fable 5.2 max · Opus 5.5 xhigh · GPT-6.1 Terra xhigh · Haiku 5.5 high | Published | 1,500 AI credits a month including the flex allotment (1 credit = $0.01).$15 of token-priced usage each month, as published. Gemini 4 Carbon is not included on Pro. Copilot plans · Copilot model pricing · Gemini 4 Carbon in Copilot |
Copilot Pro+GitHub Copilot | $39 | Fable 5.2 | 64.6 | $70 | 1.8× | 5.4Gemini 4 Carbon high · Gemini 4 Carbon medium · GPT-6.1 Terra xhigh · Haiku 5.5 high | Published | 7,000 AI credits a month including the flex allotment (1 credit = $0.01).$70 of token-priced usage each month, as published. Copilot plans · Copilot model pricing · Gemini 4 Carbon in Copilot |
Copilot MaxGitHub Copilot | $100 | Fable 5.2 | 64.6 | $200 | 2× | 6Gemini 4 Carbon high · Gemini 4 Carbon medium · GPT-6.1 Terra xhigh · Haiku 5.5 high | Published | 20,000 AI credits a month including the flex allotment (1 credit = $0.01).$200 of token-priced usage each month, as published. Copilot plans · Copilot model pricing · Gemini 4 Carbon in Copilot |
OpenCode GoOpenCode | $10 | DeepSeek V4.1 Flash 1017 | 50.4 | $100 | 10× | 0.89Gemini 4 Argon high · DeepSeek V4.1 Flash 1017 maxlittle on frontier work | Published | A monthly dollar allowance per model; 20% of it per five hours and 50% per week.Per-model monthly dollar allowances as published. Each model has its own pool; they are alternatives and do not add up. |
OpenCode Go PlusOpenCode | $40 | DeepSeek V4.1 Flash 1017 | 50.4 | $200 | 5× | 0.89Gemini 4 Argon high · DeepSeek V4.1 Flash 1017 maxlittle on frontier work | Published | A monthly dollar allowance per model; 20% of it per five hours and 50% per week.Per-model monthly dollar allowances as published. Each model has its own pool; they are alternatives and do not add up. |
GLM Coding Plan LiteClaude Code, Codex, ZCode | $18 | GLM-5.3 | 31.7 | $66–$176 | 3.7–9.8× | 0.001–0.0028GLM-5.3 max · GLM-5.3-Flashlittle on frontier work | Published | 2,000 credits per five hours and 10,000 per week.Weekly credits converted with Z.ai's published token-to-credit formula and valued at GLM API prices. Low end: every token at the least favourable rate, all peak hours. High end: most favourable rate, all off-peak (half credits). |
GLM Coding Plan ProClaude Code, Codex, ZCode | $80 | GLM-5.3 | 31.7 | $399–$1.1k | 5–13× | 0.0014–0.0038GLM-5.3 max · GLM-5.3-Flashlittle on frontier work | Published | 12,000 credits per five hours and 60,000 per week.Weekly credits converted with Z.ai's published token-to-credit formula and valued at GLM API prices. Low end: every token at the least favourable rate, all peak hours. High end: most favourable rate, all off-peak (half credits). |
GLM Coding Plan MaxClaude Code, Codex, ZCode | $168 | GLM-5.3 | 31.7 | $930–$2.5k | 5.5–15× | 0.0016–0.0042GLM-5.3 max · GLM-5.3-Flashlittle on frontier work | Published | 28,000 credits per five hours and 140,000 per week.Weekly credits converted with Z.ai's published token-to-credit formula and valued at GLM API prices. Low end: every token at the least favourable rate, all peak hours. High end: most favourable rate, all off-peak (half credits). |
Mistral Vibe ProVibe CLI and IDE | $14.99 | Mistral Large 4 | 32.6 | $140 | 9.4× | 0.063DeepSeek V4.1 Flash 1017 maxlittle on frontier work | Published | A monthly Vibe Code allowance of €127.50 and an API allowance of €12.75, both resetting on the first of the month.Mistral shows two monthly allowances in the account page: €127.50 of Vibe Code usage and €12.75 of API usage (read from a subscriber's Education plan, 12 November 2026), spendable on Mistral's own models and the GLM and DeepSeek models it hosts. Euro amounts are taken at face value as dollars. Pro carries the same allowances as the Education plan, which is Pro with a student discount. |
Mistral Vibe StudentVibe CLI and IDE | $5.99 | Mistral Large 4 | 32.6 | $140 | 23× | 0.16DeepSeek V4.1 Flash 1017 maxlittle on frontier work | Published | Pro with a student discount, for at most 12 months: a monthly Vibe Code allowance of €127.50 and an API allowance of €12.75. Mistral's FAQ gives $5.99 a month; its plan card shows $14.99.Mistral shows two monthly allowances in the account page: €127.50 of Vibe Code usage and €12.75 of API usage (read from a subscriber's Education plan, 12 November 2026), spendable on Mistral's own models and the GLM and DeepSeek models it hosts. Euro amounts are taken at face value as dollars. Open only to verified students who have not used Vibe or Le Chat before. |
Alibaba Coding Plan ProClaude Code, Codex, OpenCode | $50 | Qwen3.7 Plus | not indexed | $250 | 5× | — | Derived | 6,000 requests per five hours, 45,000 per week and 90,000 per month. A request is one model call.90,000 monthly requests valued at $0.0028 each, the per-request cost OpenCode publishes for the same model ($60 for 21,600 requests). Qwen3.7 Plus has no Intelligence Index, so this plan appears only in the usage charts. |
Kimi PlusKimi Code | $19 | Kimi K3 · About 10× K3.1 Pro's tokens per window | 46.5 | $105–$180 | 5.5–9.5× | 0.019–0.033Kimi K3.1 Prolittle on frontier work | Derived | Rolling five-hour limit and a monthly pool. Half of Pro's agent credits.Not measured. The measured Kimi Pro and Max figures (wokey.ai, November 2026) scaled by the agent-credit multiples on Moonshot's Chinese membership page (Plus 2×, Pro 4×, Max 14×, Ultra 28×), which match the measured Pro/Max ratio. The international page gives different multiples. Kimi membership · Measured limits, Nov 2026 (wokey.ai) · Kimi agent credits by tier (Chinese page) |
Kimi ProKimi Code | $39 | Kimi K3 · About 10× K3.1 Pro's tokens per window | 46.5 | $195–$345 | 5–8.8× | 0.017–0.031Kimi K3.1 Prolittle on frontier work | Measured | Rolling five-hour limit and a monthly pool. About a quarter of Max per window.Moonshot publishes no numbers. From one unofficial measurement (wokey.ai, November 2026, a reseller of these plans): tokens per five-hour window on Kimi K3.1 Pro, about 12 windows a month, valued at API prices with 93–99% cache hits. |
Kimi MaxKimi Code | $99 | Kimi K3 · About 10× K3.1 Pro's tokens per window | 46.5 | $720–$1.3k | 7.3–13× | 0.025–0.045Kimi K3.1 Prolittle on frontier work | Measured | Rolling five-hour limit and a monthly pool.Moonshot publishes no numbers. From one unofficial measurement (wokey.ai, November 2026, a reseller of these plans): tokens per five-hour window on Kimi K3.1 Pro, about 12 windows a month, valued at API prices with 93–99% cache hits. |
Kimi UltraKimi Code | $199 | Kimi K3 · About 10× K3.1 Pro's tokens per window | 46.5 | $1.4k–$2.5k | 7.2–13× | 0.025–0.045Kimi K3.1 Prolittle on frontier work | Derived | Rolling five-hour limit and a monthly pool. Twice Max's agent credits.Not measured. The measured Kimi Pro and Max figures (wokey.ai, November 2026) scaled by the agent-credit multiples on Moonshot's Chinese membership page (Plus 2×, Pro 4×, Max 14×, Ultra 28×), which match the measured Pro/Max ratio. The international page gives different multiples. Kimi membership · Measured limits, Nov 2026 (wokey.ai) · Kimi agent credits by tier (Chinese page) |
MiniMax M Plan GoMiniMax Code | $22 | MiniMax-M3.1-Flash | 29.1 | — | — | — | Not published | Five-hour and weekly windows. The quota is not published; only an unsourced community estimate exists.No absolute allowance is published, so this plan has a price and a model list but no usage figure. |
MiniMax M Plan ExploreMiniMax Code | $55 | MiniMax-M3.1-Flash | 29.1 | — | — | — | Not published | 3× the Go allowance. Five-hour and weekly windows. The quota is not published; only an unsourced community estimate exists.No absolute allowance is published, so this plan has a price and a model list but no usage figure. |
MiniMax M Plan BuildMiniMax Code | $132 | MiniMax-M3.1-Flash | 29.1 | — | — | — | Not published | 7.5× the Go allowance. Five-hour and weekly windows. The quota is not published; only an unsourced community estimate exists.No absolute allowance is published, so this plan has a price and a model list but no usage figure. |
Cursor PlusCursor Agent | $8 | Grok 4.7 | 35.5 | $35–$137 | 4.4–17× | 0.0025–0.0098Grok 4.7 xhigh · Grok 4.7 lowlittle on frontier work | Anecdotal | A weekly pool shared with the Grok apps; Grok models and Composer 2.5 only. No size is published.Cursor publishes only multiples of Plus. Range from users' usage-dashboard screenshots since the 30 October 2026 cut, which include Grok app usage, valued at Grok 4.7 prices with a 95% cache-read mix. Plus is Premium ÷ 4. Composer 2.5 is also in the pool but not benchmarked. Cursor pricing · Cursor models and pricing · User-reported pool sizes (GitHub) |
Cursor PremiumCursor Agent | $30 | Grok 4.7 | 35.5 | $138–$546 | 4.6–18× | 0.0026–0.01Grok 4.7 xhigh · Grok 4.7 lowlittle on frontier work | Anecdotal | 4× the Plus pool. Third-party models are billed at API price.Cursor publishes only multiples of Plus. Range from users' usage-dashboard screenshots since the 30 October 2026 cut, which include Grok app usage, valued at Grok 4.7 prices with a 95% cache-read mix. Plus is Premium ÷ 4. Composer 2.5 is also in the pool but not benchmarked. Cursor pricing · Cursor models and pricing · User-reported pool sizes (GitHub) |
Cursor SuperCursor Agent | $100 | Grok 4.7 | 35.5 | $1.9k–$2.4k | 19–24× | 0.011–0.013Grok 4.7 xhigh · Grok 4.7 lowlittle on frontier work | Anecdotal | 15× the Plus pool. Third-party models are billed at API price.Cursor publishes only multiples of Plus. Range from users' usage-dashboard screenshots since the 30 October 2026 cut, which include Grok app usage, valued at Grok 4.7 prices with a 95% cache-read mix. Plus is Premium ÷ 4. Composer 2.5 is also in the pool but not benchmarked. Cursor pricing · Cursor models and pricing · User-reported pool sizes (GitHub) |
Cursor UltraCursor Agent | $200 | Grok 4.7 | 35.5 | $4.1k–$4.3k | 20–22× | 0.012Grok 4.7 xhigh · Grok 4.7 lowlittle on frontier work | Anecdotal | 50× the Plus pool. Third-party models are billed at API price.Cursor publishes only multiples of Plus. Range from users' usage-dashboard screenshots since the 30 October 2026 cut, which include Grok app usage, valued at Grok 4.7 prices with a 95% cache-read mix. Plus is Premium ÷ 4. Composer 2.5 is also in the pool but not benchmarked. Cursor pricing · Cursor models and pricing · User-reported pool sizes (GitHub) |
Muse Code EverydayMuse Code | $5 | Muse Spark 1.3 | 43.6 | $27–$29 | 5.4–5.8× | 0.075–0.08Muse Wonder max · Muse Ember xhighlittle on frontier work | Derived | 10–50 prompts every five hours, plus a weekly limit.The measured High range scaled by Meta's published multiples (Everyday 1×, High 5×, this tier 1×). Contributor mode, the default, gives more and is not counted. Muse Code plans · Muse Code subscriptions · Measured multiplier, Oct 2026 (Huryn) |
Muse Code HighMuse Code | $15 | Muse Spark 1.3 | 43.6 | $134–$146 | 8.9–9.7× | 0.12–0.13Muse Wonder max · Muse Ember xhighlittle on frontier work | Measured | 5× Everyday.One independent measurement (Huryn, October 2026): weekly-meter readings against API-priced usage on Muse Spark 1.3 in standard mode, 9.3× ± 0.4 the plan price. Contributor mode, the default, gives more and is not counted. Muse Code plans · Muse Code subscriptions · Measured multiplier, Oct 2026 (Huryn) |
Muse Code PowerMuse Code | $50 | Muse Spark 1.3 | 50.1 | $536–$584 | 11–12× | 1.3–1.4Muse Wonder Full max · Muse Ember xhighlittle on frontier work | Derived | 20× Everyday. Adds Muse Wonder Full.The measured High range scaled by Meta's published multiples (Everyday 1×, High 5×, this tier 20×). Contributor mode, the default, gives more and is not counted. Muse Code plans · Muse Code subscriptions · Measured multiplier, Oct 2026 (Huryn) |
Muse Code MaximumMuse Code | $200 | Muse Spark 1.3 | 50.1 | $2.1k–$2.3k | 11–12× | 1.3–1.4Muse Wonder Full max · Muse Ember xhighlittle on frontier work | Derived | 80× Everyday. Adds Muse Wonder Full.The measured High range scaled by Meta's published multiples (Everyday 1×, High 5×, this tier 80×). Contributor mode, the default, gives more and is not counted. Muse Code plans · Muse Code subscriptions · Measured multiplier, Oct 2026 (Huryn) |
Meta One CoreMeta AI (no coding agent) | $7.99 | — | not indexed | — | — | — | Not published | More image and video generation with Meta AI. It includes no coding tool; Meta's coding plans are Muse Code.No absolute allowance is published, so this plan has a price and a model list but no usage figure. |
Meta One PremiumMeta AI (no coding agent) | $19.99 | — | not indexed | — | — | — | Not published | More image and video generation with Meta AI. It includes no coding tool; Meta's coding plans are Muse Code.No absolute allowance is published, so this plan has a price and a model list but no usage figure. |
Devin ProDevin CLI and Desktop | $20 | Fable 5.2 | 64.6 | $128–$147 | 6.4–7.3× | 11–13Fable 5.2 max · Opus 5.5 xhigh · Gemini 4 Argon high · Haiku 5.5 high | Measured | A daily and a weekly quota, drawn down at API prices. No size is published.Measured: sessions on four different models read against the weekly quota (Real API Pricing, October–November 2026) came to list-priced usage worth $29–34 a week. Fable 5.2 is assumed to draw like the Fable 5.1 that was measured. GPT usage can be billed to a ChatGPT subscription instead. Devin pricing · Devin models · Real API Pricing (measured allowances) |
Devin MaxDevin CLI and Desktop | $200 | Fable 5.2 | 64.6 | $1.8k–$3k | 8.9–15× | 15–26Fable 5.2 max · Opus 5.5 xhigh · Gemini 4 Argon high · Haiku 5.5 high | Measured | A much larger weekly quota with no daily cap. No size is published.Measured: two sessions read against the weekly quota (Real API Pricing, September 2026), on GPT-6 Astra ($410 a week) and Opus 5.5 ($701 a week). The two disagree, so the range is wide. Devin pricing · Devin models · Real API Pricing (measured allowances) |
