כמה באמת יעלה לכם לעבוד עם Copilot Cowork? הכירו את הכלי שיעזור לכם לתכנן מראש

עם המעבר של Copilot Cowork  ל-GA, מיקרוסופט משנה את הדרך שבה ארגונים מתכננים את עלויות ה-AI שלהם.

במקום מודל תשלום קבוע מבוסס משתמש (USL – User License Seat) , הרישיונות פוצלו:

לצד רישיון ה-Copilot 365 המהווה דרישת בסיס, השימוש ביכולות מתקדמות ומורכבות כמו Cowork מתבסס על Copilot Credits, בהתאם לסוג הפעולה, רמת המורכבות, שימוש בכלים חיצוניים וזמן הריצה.

כדי להבין "כמה זה עולה לנו" פיתחנו ב-UBTECH כלי שמסייע לארגונים לקבל תמונה ברורה יותר של העלויות הצפויות, עוד לפני שמתחילים להרחיב את השימוש בארגון.

הכלי שעושה סדר בצריכת קרדיטים

כדי להפוך את חוסר הוודאות לתהליך תכנוני ברור, יצרנו עבורכם את ה-Cowork Credit Estimator. המטרה כאן, לעזור בחיזוי עלות הצריכה העתידית עם Cowork בעזרת נתוני אמת הקיימים כבר היום בממשק האדמין הארגוני (כגון כמות משתמשים, מספר אינטראקציות ורמות אימוץ בפועל).

כדי להעניק לארגון הבנה טובה יותר של התקציב, מודל התמחור מאפשר לבחור בין שני מסלולי התקשרות:

  • צריכה חודשית שוטפת (Pay as you Go): תשלום בהתאם להיקף השימוש בפועל באותו החודש.
  • רכישת חבילת קרדיטים מראש (Pre-Purchase  Credit – P3) : רכישת בנק קרדיטים מוגדר שנתי מראש לניהול תקציבי יציב, מבוקר עם הנחות לפי מסלול הרכישה (Tier’s)

UBTECH · Copilot Cowork Credit Estimator

Copilot Cowork Credit Estimator

Model how many Copilot Credits your organization may consume with Microsoft 365 Copilot Cowork. Enter who will use Cowork and how intensively, then see an estimated monthly credit and budget view. Provided for illustrative budgeting only — validate against your own usage.

⚡ Estimates assume Anthropic Opus 4.8 · List pricing: 100 Copilot Credits = $1.00 USD
📘

Model guidance

Cowork is billed on usage, in Copilot Credits. The price of each task is built from four inputs — the more each one is exercised, the more credits a task consumes. Use this guidance to gauge how light, medium, or heavy your team’s prompts are likely to be before filling in the steps below.

What drives the cost of a task
🧠

Models

The AI model chosen for each task — quality, speed, and cost vary by what the task demands.

🗂️

Context

Understanding of the people, roles, and collaboration behind the work — emails, files, meetings, and past interactions.

🛠️

Tools

Actions the system takes to get work done: sending emails, scheduling meetings, updating documents, and more.

⏱️

Runtime

Managed cloud orchestration that runs agents and keeps them working across tasks, including long-running work.

Light, medium & heavy tasks

Light

~70–200 credits

Narrow context, a lightweight model, 0–1 tool calls, minimal runtime — producing 0–1 deliverables.

“Create a weekly status update for my team with my top priorities and key meetings, saved as a draft.”

Medium

~400–600 credits

Richer context, a capable model, several tool calls, moderate runtime — producing 2+ outputs.

“Prepare me for a customer meeting: pull relevant emails, calendar, CRM, and files into a briefing doc.”

Heavy

~1,500+ credits

Broad context aggregation, a high-quality model, many tool calls, sustained runtime — many outputs.

“Analyze 6 months of exported product usage data, classify patterns, and produce a leadership-ready analysis.”

Credit ranges are illustrative and assume Anthropic Opus 4.8; actual consumption varies with task complexity. You can fine-tune the per-prompt credit values in Step 3.

1

Who will use Cowork?

Enter the number of users in each role that will work with Cowork.

User groupNumber of users
Corporate Knowledge Workers
Customer-Facing Knowledge Workers
Technical Workers
Managers & Senior Leaders
Total users0
2

How many light, medium & heavy prompts per user each month?

Estimate the monthly prompt mix for a typical user in each role. The per-user credit figure updates automatically.

User group Light Medium Heavy Credits / user / mo
Corporate Knowledge Workers 0
Customer-Facing Knowledge Workers 0
Technical Workers 0
Managers & Senior Leaders 0
Light — narrow context, lightweight model, 0–1 tool calls, minimal runtime (0–1 deliverables) Medium — richer context, capable model, several tool calls, moderate runtime (2+ outputs) Heavy — broad context, high-quality model, many tool calls, sustained runtime (many outputs)
3

How many Copilot Credits are consumed per prompt?

Default rates reflect Microsoft Frontier customer usage (as of 27 May 2026). Adjust to match your organization.

Light prompt

Narrow context, lightweight model, 0–1 tool calls, minimal runtime — 0–1 deliverables.

Medium prompt

Richer context, capable model, several tool calls, moderate runtime — 2+ outputs.

Heavy prompt

Broad context aggregation, high-quality model, many tool calls, sustained runtime — many outputs.

4

Your estimate

Est. monthly Copilot Credits
0
across all users & roles
Est. Annual Copilot Credits
0
monthly credits × 12 months
Est. monthly Cowork budget
$0.00
list pricing · 100 credits = $1
Avg. price per user / month
$0.00
budget ÷ total users
Annual Cost ($)
$0.00
monthly budget × 12 months
Total users: 0 Total monthly prompts: 0
🏷️
Recommended Copilot Credit Pre-Purchase Plan (P3)
Enter usage above to see your recommended plan.
Copilot Credit Pre-Purchase Plan (P3) pricing tiers — Tier, Copilot Credits and Discount
P3 pricing tiers — Copilot Credits Guide (June 2026).

Helpful resources

Learn more about managing Cowork experiences and Copilot Credits.

1. Per-prompt credit disclaimer. The information presented herein is derived from aggregated and anonymized data collected from customers who participated in the Cowork Frontier program. It is intended for illustrative and informational purposes only and may not reflect the experience or results of all customers. It should not be relied upon as definitive, complete, or predictive of future outcomes.

2. Estimate disclaimer. This estimator is designed to help inform budgeting considerations for potential Cowork purchases. The data represents generalized estimates and may vary based on individual customer scenarios. Microsoft does not guarantee the accuracy of this information and disclaims any warranties or assurances related to expected results or outcomes.

Estimates assume Anthropic Opus 4.8. List pricing shown (100 Copilot Credits = $1.00 USD); your contracted pricing may differ. This tool is provided by UBTECH Solutions LTD. to support customer planning and is not an official Microsoft pricing quote.

Powered By Nati Papirovitch, Head of AI & Business Solutions