Team Plan Overview

Product Introduction

Coding Flow Team Edition is an AI coding subscription service designed for development teams. It uses Credits as a unified metering unit, allowing a single subscription to cover multiple developers. It supports flexible multi-model switching and provides team management capabilities including a management console, seat allocation and revocation, and member usage analytics.

Key Features:

• Flexible Multi-Model Switching: Switch between models on demand, with unified Credits deduction.
• Compatible with Multiple Tools: Works with Claude Code, Codex, Gemini CLI, OpenClaw, OpenCode, and other mainstream coding tools and popular Agent tools.
• Multiple Pricing Tiers: Standard and Professional tiers to match different usage intensities.
• Team Management: Management console with seat allocation/revocation and member usage analytics.
• Budget Control: Weekly quota refresh, hard stop on exhaustion, zero surprise bills.
• Data Security: Commitment not to retain conversation data, meeting enterprise-level data privacy requirements.

Note The rights and benefits of the team version and the individual version are independent of each other and

can be activated and used simultaneously. The quotas for both are calculated and consumed separately, without affecting each other.

Comparison with Individual Plan

Dimension	Individual Plan	Team Plan
Use Case	Individual developers	Teams / Enterprise daily development
Billing	Monthly payment	Per-seat monthly payment, each seat includes independent Credits quota
Metering Unit	Model call count	Credits
Quota Ownership	Shared across all Keys under user	Per-seat independent, no cross-impact
Time Window	5h + 7-day dual window	5h + 7-day dual window
Exhaustion Handling	Paused when quota exhausted, wait for window refresh	Can use team Credits Pool to top up without interruption
API Key	sk-apv-cp-xxx	sk-apv-cp-team-xxx
Management	Self-managed	Owner manages seats / members / billing
Model Updates	Standard models, regular updates	Latest models available, priority access to new releases

What is a "Seat"?

A seat is the unit of access in the Team Plan — it is not a separately purchased product. When a team subscribes to a plan, the number of seats determines how many members can use the service simultaneously.

In simple terms:

• Team subscription = one plan × N seats, a single payment covers the entire team
• 1 seat = 1 member's access, including independent Credits quota and a dedicated API Key
• Seats are managed centrally by the team Owner — they can be assigned to members or revoked and reassigned to others

For example: A team purchases the Standard tier with 2 seats ($129 × 2 = $158/month), enabling 2 members to use the service simultaneously, each with an independent quota of 13 million Credits/week. No individual payments needed — the team subscribes and manages everything centrally.

Plans & Pricing

Team Plan Tiers

Tier	Price / Month	Annual Total (30 days bonus, 13 months)	Quota / Week / Seat	Quota / 5 Hours / Seat	Use Case
Basic	$129 / seat / month	$1,548 / seat / year	13M Credits	8M Credits	Small-medium team daily development
Flagship	$249 / seat / month	$2,988 / seat / year	26M Credits	16M Credits	Heavy usage / Large projects
Flagship	$499 / seat / month	$5,988 / seat / year	54M Credits	33M Credits	Agent, complex tasks, long context, or high-concurrency teams

警告

• Annual Payment Promotion: Each seat receives a 30-day bonus upon subscription, totaling 13 months of quota.
• Basic Edition: Each seat per month receives a fixed bonus of 4M credits shared pool, automatically added.
• Professional Edition: Each seat per month receives a fixed bonus of 8M credits shared pool, automatically added.
• Flagship Edition: Each seat per month receives a fixed bonus of 18M credits shared pool, automatically added.

• Minimum 2 seats required for Team Plan subscription
• Additional seats can be purchased anytime, prorated by remaining cycle

Credits Pool Top-up Packs

Credits Top-up Packs

Top-up Pack Tier	Credits / Pack	Original Price / Pack	Discounted Price / Pack
Basic Pack	50M	$250	$200 (20% off)
Advanced Pack	200M	$1,000	$750 (25% off)
Flagship Pack	500M	$2,500	$1,750 (30% off)
Custom Pack	500M+ (default)	-	The amount displayed on the page shall prevail

• Purchasing Top-up Packs: Click "Team Shared Credits Pack" in the billing section to purchase. After purchase, the Credits Pool is enabled by default.
• Top-up Pack Validity: Valid for 365 days by default; only usable during an active seat subscription period (if the team plan expires, the top-up pack becomes unavailable; it automatically resumes when the team plan is re-subscribed).
• Per-Seat Top-up Cap: By default, each account can use the full Credits Pool. The Owner can set a per-seat top-up cap under Members & Seats to prevent a single person from consuming the entire pool.

Subscription Management

Operation	Rule
Add Seats	Available anytime; new seats share existing subscription expiry. Fee = Seat price × Quantity ÷ Total days in cycle × Remaining days
Upgrade Seat	Click upgrade on the seat in subscription details to upgrade from a lower tier to a higher tier, paying the price difference
Reassign Seat	Old member's used quota stays in current period (non-refundable); new member uses seat's remaining / new cycle quota
Remove Seats	Not supported directly; simply don't renew at expiry
Pre-purchase Next Plan	While the current plan is active, you can purchase the next billing cycle in advance. Validity is added directly upon successful purchase: new expiry date = original expiry date + purchased duration. Payment extends service continuously. The maximum stackable duration is 12 months
Refund	Digital goods are non-refundable

Seat Add-on Fee Calculation Formula: Add-on fee = Seat price × Number of seats ÷ Total days in cycle ×

Remaining days

Example: Standard tier ($129/seat/month), current cycle is 30 days total with 20 days remaining, adding 1 seat: Add-on fee = $129 × 1 ÷ 30 × 20 = $86

Credits Billing Mechanism

Billing Description

Credits deducted per call = sum of (each token type × the corresponding model coefficient).

Deduction Order

Seat quota → gifted Credits → paid packs

Calculation Formula

• Credits consumption: input tokens × input coefficient + cache write tokens × cache write coefficient + cache read tokens × cache read coefficient + output tokens × output coefficient.
• Coefficients: based on each model's official pricing. Coefficients are published when new models launch and updated in sync with official price changes (coefficient = the item's official unit price ÷ 5).

Coefficients

Model	Input	Cache Write	Cache Read	Output
claude-opus-5-5	0.8	1	0.04	4
claude-opus-5	1	1.25	0.1	5
claude-opus-4-8	1	1.25	0.1	5
claude-opus-4-7	1	1.25	0.1	5
claude-sonnet-5-5	0.4	0.5	0.04	2
claude-sonnet-5	0.4	0.5	0.04	2
claude-haiku-5-5	0.1	0.125	0.01	5
gpt-6-astra	2	2.5	0.2	10
gpt-6-sol	0.4	0.5	0.04	2
gpt-6-luna	0.02	0.025	0.002	0.1
gpt-5.6-luna	0.04	0.05	0.004	0.24
gpt-5.6-terra	0.4	0.5	0.04	2.4
gpt-5.6-sol	1	1.25	0.1	6
gpt-5.5	1	0	0.1	6
glm-5.2	0.23	0	0.0576	0.8
deepseek-v4-pro	0.264	0	0.0088	0.792
kimi-k3	0.6	0	0.06	3

Calculation Example

Using claude-opus-4-8 as an example, the estimated per-request consumption breakdown is as follows (different models have different unit prices; actual consumption is based on the billing statement):

Token Type	Tokens	Credits Consumed
Input tokens	225 Tokens	225 Credits
Output tokens	1,121 Tokens	5,605 Credits
Cache write tokens	1,057 Tokens	1,321.25 Credits
Cache hit tokens	61,225 Tokens	6,122.5 Credits
Total		13,273.75 Credits

Quota Mechanism · Dual-Layer Time Windows

Window	Type	Reset Logic
5-Hour	Sliding Window	Each call occupies 5h; released one by one after 5h
7-Day	Rolling Window	Each call consumes 7 days, released one by one after 7d

Key Rules:

• Both windows are active simultaneously; service is blocked when either is exhausted
• When blocked, the UI displays the specific reset time (e.g., "Quota will reset at 14:32")
• Each member has an exclusive quota for their API Key, and the quota is independently blocked when exhausted

Quota Details

What do "5-Hour Quota" and "7-Day Quota" mean?

Coding Flow Team Plan uses a dual-layer sliding window mechanism to manage Credits quota:

Dual Window Quota (Standard Tier Example)

Window	Meaning	Standard Quota/Seat	Professional Quota/Seat	Flagship Quota/Seat
5-Hour	Maximum Credits consumable within any consecutive 5 hours	8M	16M	33M
7-Day	Maximum Credits consumable within any consecutive 7 days	13M	26M	54M

Both windows are active simultaneously; service is blocked when either is exhausted. Each member's quota is calculated independently and does not affect others.

---

How are Credits consumed? What's different from the Individual Plan?

Individual Plan vs Team Plan Billing Comparison

Dimension	Individual Plan	Team Plan
Metering Unit	Request count (1 call = 1, model-agnostic)	Credits (dynamically calculated by model × token)
Model Impact	All models consume equally (1 call each)	Different model coefficients; premium models consume more
Token Impact	No impact (any token count = 1 call)	More tokens = higher consumption
Quota Ownership	Shared across all Keys under user	Per-seat independent, no cross-impact
Management	Self-managed	Owner manages centrally, can view member usage

---

Is there a big difference in consumption between models?

Yes. The Team Plan bills by Credits, and different models have different coefficients:

• Premium models (e.g., claude-opus-4-8): Higher output coefficient, consuming more Credits per request
• Lightweight models (e.g., gpt-5): Lower coefficient, consuming less for the same task

We recommend choosing the appropriate model based on task complexity: use lightweight models for simple tasks to save quota, and premium models for complex tasks to ensure quality.

---

Can I specify which model to use?

Yes. Coding Flow Team Plan supports all available models, and members can switch freely:

Tool	How to Switch
Claude Code	/model command
Codex	/model command
Kilo Code	Settings panel
Other tools	Refer to each tool's model configuration docs

---

Is context length limited?

No. Coding Flow Team Plan does not impose additional context length limits. You can fully utilize each model's maximum context window.

However, note that longer context means more token consumption, and correspondingly higher Credits consumption.

---

Are there usage scenario restrictions?

Quota is limited to use within designated AI coding tools, but there are no restrictions on usage scenarios within the tools:

Scenario Type	Supported
Code writing & generation	✅
Code debugging & fixing	✅
Task breakdown & planning	✅
PRD analysis & requirement understanding	✅
Test case generation	✅
Code review & refactoring suggestions	✅
Documentation & comment generation	✅
Architecture design & technical discussions	✅
Other lawful & compliant scenarios	✅

As long as requests are initiated within supported coding tools — whether writing code, making plans, or generating test cases — they fall within normal usage.

---

How to estimate Team Plan usage?

Using the Standard tier (13M Credits/week/seat) with the claude-opus-4-8 model as an example:

Usage Estimate (Standard · claude-opus-4-8)

Scenario	Per-Request Consumption (approx.)	Weekly Available Calls (approx.)
Simple Q&A (short context)	~5,000 Credits	~2,600 calls
Regular coding tasks	~13,000 Credits	~1,000 calls
Complex tasks (long context)	~50,000 Credits	~260 calls

The above are reference estimates. When using lightweight models (e.g., gpt-5), the same token consumption costs fewer Credits, allowing more available calls.

Supported Models

Coding Flow Team Plan supported models. For detailed model information, see Model List.

Plan	Supported Models
Standard Edition	claude-opus-5-5 claude-opus-5 claude-opus-4-8 claude-opus-4-7 claude-sonnet-5-5 claude-sonnet-5 claude-haiku-5-5 gpt-6-sol gpt-6-astra gpt-6-luna gpt-5.6-luna gpt-5.6-terra gpt-5.6-sol gpt-5.5 glm-5.2 deepseek-v4-pro kimi-k3
Professional Edition	claude-opus-5-5 claude-opus-5 claude-opus-4-8 claude-opus-4-7 claude-sonnet-5-5 claude-sonnet-5 claude-haiku-5-5 gpt-6-sol gpt-6-astra gpt-6-luna gpt-5.6-luna gpt-5.6-terra gpt-5.6-sol gpt-5.5 glm-5.2 deepseek-v4-pro kimi-k3

Quick Start

Three steps to subscribe and connect to Coding Flow Team Edition: choose a plan, get API Key, configure AI tools.

1

Subscribe to Team Plan

1. Log in to Console
2. Navigate to the Plans page and select a Team Plan tier
3. Choose the number of seats and complete payment
4. After payment, your team is automatically created and you become the Team Owner

2

Invite Members and Assign Seats

1. Enter the Team Management console
2. The Team Owner navigates to the "Members & Seats" page
3. Click "Invite Member" (only Member role can be invited by default), and configure options such as invite link expiration, maximum number of invitees, and auto-assign seats
4. After clicking "Generate Link", share the link with team members
5. Team members click the invite link to log in and can choose to accept the invitation (if a team member is not yet registered in the system, they will automatically join the team after registration without needing confirmation)
6. Once accepted, the seat is automatically activated and the system generates a team-exclusive Key for the member

3

Configure AI Tools

After members receive their team-exclusive Key (sk-apv-cp-team- prefix), configure it in their AI tool to start using.

For more tool configuration details, refer to Tool Integration.

Contact Us

Coding Flow Community Group

Scan QR code to join via Feishu (Lark)