Usage Rates Ka Vistrit Vivaran

Yeh page vistar se batata hai ki Neotask usage-based charges kaise calculate karta hai. Individual plan par adhikansh users ko yeh charges kabhi nahi dikhenge. Yeh vivaran advanced enterprise users ke liye hai jinke agents scale par operate karte hain.

Plans ka overview aur billing kaise kaam karti hai ke liye, Billing aur Plans dekhein.

Usage Kaise Calculate Hota Hai

Har baar jab aapka agent message process karta hai, woh AI tokens (input aur output) consume karta hai. In tokens ki cost is par nirbhar karti hai ki aapka agent kaunsa AI model use karta hai. System char prakar ke tokens ko alag-alag track karta hai, har ek ki apni rate hai:

Input tokens: model ko bheja gaya prompt aur context
Output tokens: model dwara generate kiya gaya response
Cache read tokens: prompt cache se serve kiye gaye tokens (input se saste)
Cache write tokens: bhavishy ke punar upyog ke liye prompt cache mein likhe gaye tokens

Token Cost Formula

Cost = (input tokens / 1,000,000) x input rate
     + (output tokens / 1,000,000) x output rate
     + (cache read tokens / 1,000,000) x cache read rate
     + (cache write tokens / 1,000,000) x cache write rate

Source: server/src/config/modelPricing.ts

AI Model Ke Anusaar Token Rates

Yeh har supported AI model ke liye per-token rates hain. Sabhi rates USD per 1 million tokens mein hain.

Anthropic Claude

Model	Input	Output	Cache Read	Cache Write
Claude Opus 4.6	$15.00	$75.00	$1.50	$18.75
Claude Sonnet 4.5	$3.00	$15.00	$0.30	$3.75
Claude Haiku 4.5	$0.80	$4.00	$0.08	$1.00
Claude 3.5 Sonnet	$3.00	$15.00	$0.30	$3.75
Claude 3.5 Haiku	$0.80	$4.00	$0.08	$1.00
Claude 3 Opus	$15.00	$75.00	$1.50	$18.75
Claude 3 Sonnet	$3.00	$15.00	$0.30	$3.75
Claude 3 Haiku	$0.25	$1.25	$0.03	$0.30

OpenAI

Model	Input	Output	Cache Read	Cache Write
GPT-4o	$2.50	$10.00	$1.25	$2.50
GPT-4o Mini	$0.15	$0.60	$0.075	$0.15
GPT-4 Turbo	$10.00	$30.00	$5.00	$10.00
GPT-4	$30.00	$60.00	$15.00	$30.00
o1	$15.00	$60.00	$7.50	$15.00
o1-mini	$3.00	$12.00	$1.50	$3.00
o3-mini	$1.10	$4.40	$0.55	$1.10

Google Gemini

Model	Input	Output	Cache Read	Cache Write
Gemini 2.0 Flash	$0.10	$0.40	$0.025	$0.10
Gemini 1.5 Pro	$1.25	$5.00	$0.3125	$1.25
Gemini 1.5 Flash	$0.075	$0.30	$0.01875	$0.075

Source: server/src/config/modelPricing.ts lines 8-42

Platform Fees aur Markups

Kai alag fees hain jo laagu ho sakti hain is par nirbhar karke aapka agent kaise use ho raha hai. Yeh ek hi fixed fee nahi hain; har ek alag situations mein laagu hoti hai.

1. Credits Mode Markup (System Key): 20%

Neotask ki managed API keys (System Key mode) use karte waqt, raw token cost par 20% markup laagu hota hai. Yeh API key management, automatic failover, model routing aur infrastructure cover karta hai.

Agar aap BYOK mode (Bring Your Own Keys) use karte hain, toh Token costs seedha aapke AI provider ko jaati hain, lekin aapke credits se 20% platform fee laagu hoti hai.

Source: server/src/config/modelPricing.ts line 53, CREDIT_MODE_MARKUP_PCT = 0.20

2. Overage Platform Fee (Variable)

Jab aapka usage included credit pool se zyada hota hai, toh bachi cost overage ban jaati hai. Overage amount par platform fee laagu hoti hai. Fee aapke plan level par nirbhar karti hai:

Level	Overage Fee
Individual	25%
Business	15%
Enterprise	10%

Overage fee sirf us amount par charge hoti hai jo aapke credit pool se zyada hai. Pool ke andar usage par koi additional fee nahi hai.

Overage automatically aapke card par tab charge hota hai jab unsettled amount $10 tak pahunch jaata hai (minimum Stripe charge: $0.50).

Source: server/src/config/planConfig.ts, overageFeePct aur overageChargeThreshold

3. Automation Markup: 50% Se Lekar 33% Tak Progressive

Jab aapke agents automated jobs (scheduled tasks, cron jobs, recurring automations) execute karte hain, toh token cost par additional automation markup laagu hota hai. Yeh isliye kyunki automated agents manual kaam ki jagah lete hain. Woh bina dhyan ke, schedule par chalte hain, un tasks ko handle karte hain jo nahi toh employees ki zaroorat hoti.

Automation markup progressive brackets (income tax ki tarah) use karta hai. Aap automated usage ke pehle dollars par zyada rate pay karte hain aur jaise aapka automated spend badhta hai kam rate. Koi step effects nahi; har dollar sirf uski bracket ki rate par charge hota hai.

Accumulated Automated Spend (per billing cycle)	Marginal Markup Rate
$0 - $10	50%
$10.01 - $25	45%
$25.01 - $50	40%
$50.01 - $100	37%
$100.01+	33%

Source: server/src/config/automationMarkup.ts lines 38-44, DEFAULT_AUTOMATION_BRACKETS

4. Coding Task Markup: Fixed 50%

Automated jobs jo coding tools (file reads, writes, code execution, bash commands) involve karte hain, un par fixed 50% markup lagta hai. Progressive reduction coding tasks par laagu nahi hoti; yeh hamesha 50% hai chahe accumulated spend kitna bhi ho.

Coding tools: exec, read, write, bash, code

Source: server/src/config/automationMarkup.ts line 49, CODING_TASK_MARKUP_RATE = 0.50

Costs Kaise Accumulate Hote Hain

Yeh fees situation ke anusaar accumulate ho sakti hain:

Interactive chat (System Key mode):

Raw token cost + 20% credits mode markup
Koi automation markup nahi (automated nahi hai)

Interactive chat (BYOK mode):

Raw token cost + credits se 20% platform fee

Automated cron job (System Key mode):

Raw token cost + 20% credits mode markup + automation markup (50%-33%)

Automated coding task (System Key mode):

Raw token cost + 20% credits mode markup + 50% coding task markup

Automation Billing Example

Ek scheduled cron job daily chalta hai. AI agent ko apna kaam complete karne mein lagbhag 2-3 ghante lagte hain. Job $10 raw tokens mein consume karta hai (AI provider ki actual cost).

Breakdown:

Token cost: $10 (AI provider ko pass-through; yeh hai compute ki actual cost)
Automation markup: $10 x 50% = $5 (bina attention ke kaam chalane ki platform fee)
Total charge: $15

$10 actual AI compute cover karte hain jo aapke agent ne consume kiya. $5 automation fee hai. Platform ne kaam schedule par chalaya, execution monitor kiya, retries handle kiye aur results deliver kiye, sab bina kisi ko computer par hona zaroorat ke.

Automation zyada kyun cost karta hai: Automated agents insaani kaam ki jagah lete hain. Ek cron job jo har subah aapki analytics check karta hai, reports draft karta hai, inventory monitor karta hai ya incoming orders process karta hai, woh aisa kaam kar raha hai jo nahi toh ek vyakti ka samay chahta. Automation markup us unattended execution ki value reflect karta hai, aur jitna zyada aap automate karte hain utna kam hota jaata hai, scale ko reward karte hue. Ek tenant jo $100+/cycle automated jobs par spend karta hai sirf 33% markup pay karta hai 50% ki jagah.

Top-Up Credits

Aap usage ke liye pre-pay karne ke liye kabhi bhi additional credits khareed sakte hain:

Minimum top-up: $5
Maximum top-up: $10,000
Credits kabhi expire nahi hote. Woh aapke account mein tab tak rehte hain jab tak use na ho jaayein.
Auto top-up: Vaikalpik roop se, jab aapka balance ek threshold se neeche gire tab automatic top-ups configure karein (default: $2 balance $10 top-up trigger karta hai)

Top-up credits credit pool khatam hone ke baad aur overage accumulate hone se pehle consume hote hain. Iska matlab top-ups aapke included credits aur overage charges ke beech buffer ka kaam karte hain.

Source: server/src/services/overageCharger.ts, server/src/services/balanceService.ts

Credit Deduction Order

Jab aapka agent ek task complete karta hai, cost is order mein deduct hoti hai:

Signup credits (ek baar ke $10 signup credits, reset nahi hote)
Top-up balance (khareedey gaye credits, kabhi expire nahi hote)
Overage (aapke card par applied platform fee ke saath charge hota hai)

Iska matlab aapke included credits hamesha pehle use hote hain, phir koi bhi khareedey gaye top-ups, aur dono khatam hone ke baad hi overage billing shuru hoti hai.

Source: server/src/services/balanceService.ts lines 240-286

Budget Controls

Enterprise users ke paas built-in budget enforcement hai:

Global daily budget: Default $500/din (configurable). Exceed hone par, agent gateway runaway costs rokne ke liye automatically shut off ho jaata hai.
Per-agent spend limits: Har individual agent ke liye daily limit set karein. Jab agent apni limit tak pahunchta hai, uski sessions automatically pause ho jaati hain.
Automatic shutdown: Configurable. Budget exceed par gateway ka automatic shutdown on ya off karein.
Real-time enforcement: Budget har 30 seconds mein check hota hai.

Source: Neotask-Electron/src/main/budgetEnforcer.ts, Neotask-Electron/src/main/agentSpendLimitStore.ts

Apna Usage Dekhna

Desktop app mein Usage page par yeh sab real-time mein track karein:

Model, provider, agent, channel aur date ke anusaar token-level breakdown
Cost breakdown: input, output, cache read aur cache write costs alag se dikhaye jaate hain
Time granularity: today, last 24h, 7 days, 30 days, 365 days ya custom range
Per-agent attribution: dekhein ki kaunsa agent kya consume kar raha hai
Per-channel attribution: messaging channel ke anusaar cost breakdown
Per-provider attribution: AI provider ke anusaar cost (Anthropic, OpenAI, Google)
External analysis ke liye CSV/JSON export

Apni subscription manage karne ke baare mein zyada ke liye Billing aur Plans dekhein.