🎧 MSPs

AI cost and governance your SMB clients can afford

For managed service providers. A lighter service across your client base, and a clear upgrade path when a client builds a serious AI workload.

Is this you?

Your clients use Copilot, ChatGPT and AI features in their SaaS
They need simple rules and a monthly report, not a governance program
Some will build an AI agent or voice line and need more

Priced on governed scope, not on consumption. You pay for the workloads AICost governs. No token markup, no share of your AI bill and no share of savings, so our price never grows because your usage does.

Plans

MSPCost AI Portfolio

$1,500/mo

Traditional MSPs serving SMB clients

Basic AI cost and governance for your first 10 client tenants
Add more clients at a flat price each
Multi-client view and reports

10 client tenants included; $99 per extra client a month.
Typical resale per client a month: $249 to $499

Talk to us about MSPCost

Includes

AI services and subscriptions found per clientSpend reviewBasic usage policiesCopilot, vendor and API visibilityObvious risks and wasteMonthly watch and report

When a client outgrows it: When a client builds a RAG system, an agent, a voice line or AI document processing, move that workload to an AICost Workload Review at partner price.

MSPCost, the MSP suite ↗

An illustrative workload, priced by AICost engines

What AICost changes in one AI workflow

Each step below is priced by its own AICost engine at the volume shown. The levers and the governance envelope are what AICost adds. Numbers refresh with model prices.

AI front desk: phone, scheduling, intake and eligibility

6,000 patient call a month · 7 steps

AI voice agent answers inbound calls after hours or on overflow, then Identify the patient and intent (schedule, refill, billing, directions), then Book or reschedule in the PM system; send reminders and intake forms, then Verify insurance eligibility.

Price each call $2,196 $1,686/mo ↓ 7% prompt caching↓ 12% context compaction↓ 4% model routing Voice Agent

Limit turns per call $3,266 $1,765/mo ↓ 46% loop cap Agent Loop

Scheduling and EHR lookups $1,472/mo Tool Integrations

Identity checks, no clinical advice $15/mo ✓ Guardrail screening Safety Screening

Hand off to staff $3,000 $1,600/mo ↓ 47% better grounding Human Review

TCPA and AI-voice disclosure $3,654/mo ✓ Compliance evidence Compliance Cost

Cost per call vs front-desk staff /mo Support Deflection ROI

Without AICost

$13,603/mo

$2.27 per patient call

Spend shows up on next month's invoice
No per-run or daily ceiling on agents
Controls written down, not enforced
No evidence trail for audits

With AICost

$10,192/mo

$1.70 per patient call 25% lower cost

Monthly budget$2,000

Daily cap$87

Per-run ceiling$1.46

Turn cap per run16

Guardrail screening
Human review
Compliance evidence
Cost firewall at the gateway
Data-class model allowlist

Where the savings come from

🎧 Price each call$2,196 → $1,686/mo Prompt caching using the AICost Prompt Caching engine −$150 (7%)
Repeated instructions and documents are cached, so you pay less for the same prompt text. *; Context compaction using the AICost Context Trim engine −$269 (12%)
Long histories are trimmed each turn, so every call carries less text. *; Model routing using the AICost Model Router engine −$91 (4%)
Routine steps run on a cheaper model; critical steps stay on the primary model. *

🎧 Limit turns per call$3,266 → $1,765/mo Loop cap using the AICost Agent Loop engine −$1,502 (46%)
A ceiling on turns and tokens per agent run, so a stuck agent cannot run up the bill.

• Hand off to staff$3,000 → $1,600/mo Better grounding using the AICost Human Review engine −$1,400 (47%)
Answers come only from your approved sources, so fewer cases go to a person. *

What it means against the AICost price: Monthly savings on this one workload cover the MSPCost AI Portfolio fee 2.3 times.

Illustrative: one workload at the volume shown, priced from current public model prices. Your numbers come from your own workload. Open this workflow in the live demo →

Questions

What do clients pay?

You set the price. The resale range shown is a guide.

What if a client outgrows it?

Their serious workload moves to AICost at partner price; the rest stays on the MSP service.

What counts as a workload?

One AI application or business process with its own outcome and governance boundary. Steps inside it, such as models, agents and tools, are not counted separately.

Do you charge a percentage of AI spend?

No. AICost is priced on the workloads it governs. There is no token markup and no share of savings.

Other paths

Talk to AICost

Bring one workload. We price it step by step and show what changes with AICost.

Book a free consult See the live demo

Vendor / Model	Field	Why it’s inferred
Anthropic — Claude Sonnet 4.6	`cachedInput`	Derived at 10% of input rate — Anthropic publishes 90% cache-hit discount on this tier.
Anthropic — Claude Sonnet 4.5	`cachedInput`	Derived at 10% of input rate; same 90% cache-hit convention as Sonnet 4.6.
Anthropic — Claude Sonnet 4.5	`batchInput`	Derived at 50% of standard input — Anthropic documents uniform 50% Batch discount.
Anthropic — Claude Sonnet 4.5	`batchOutput`	Derived at 50% of standard output — Anthropic documents uniform 50% Batch discount.
Anthropic — Claude Haiku 4.5	`cachedInput`	Derived at 10% of input rate — Anthropic 90% cache-hit discount convention.
OpenAI — GPT-5.4 Mini	`cachedInput`	Derived at 10% of input — OpenAI documents automatic 90% discount on cache hits across GPT-5.x tier.
OpenAI — GPT-5.4 Nano	`cachedInput`	Derived at 10% of input — OpenAI 90% cache-hit convention.
OpenAI — GPT-5.4 Nano	`batchInput`	Derived at 50% of input — OpenAI Batch API uniform 50% discount.
OpenAI — GPT-5.4 Nano	`batchOutput`	Derived at 50% of output — OpenAI Batch API uniform 50% discount.
OpenAI — GPT-5.4 Pro	`cachedInput`	Derived at 10% of input — OpenAI 90% cache-hit convention.
OpenAI — GPT-5.4 Pro	`batchInput`	Derived at 50% of input — OpenAI Batch API uniform 50% discount.
OpenAI — GPT-5.4 Pro	`batchOutput`	Derived at 50% of output — OpenAI Batch API uniform 50% discount.
OpenAI — GPT-5.2	`cachedInput`	Derived at 10% of input; no residency uplift.
OpenAI — GPT-5.2	`batchInput`	Derived at 50% of input.
OpenAI — GPT-5.2	`batchOutput`	Derived at 50% of output.
OpenAI — GPT-5	`cachedInput`	Derived at 10% of input.
OpenAI — GPT-5	`batchInput`	Derived at 50% of input.
OpenAI — GPT-5	`batchOutput`	Derived at 50% of output.
OpenAI — GPT-5.5 Pro	`cachedInput`	Derived at 10% of input — OpenAI does not publish a cached rate for *-pro models; using the family convention.
OpenAI — GPT-5.5 Pro	`batchInput`	Derived at 50% of input.
OpenAI — GPT-5.5 Pro	`batchOutput`	Derived at 50% of output.
OpenAI — GPT-5.2 Pro	`cachedInput`	Derived at 10% of input — pro-tier convention.
OpenAI — GPT-5.2 Pro	`batchInput`	Derived at 50% of input.
OpenAI — GPT-5.2 Pro	`batchOutput`	Derived at 50% of output.
OpenAI — GPT-5.1	`batchInput`	Derived at 50% of input.
OpenAI — GPT-5.1	`batchOutput`	Derived at 50% of output.
OpenAI — GPT-5 Pro	`batchInput`	Derived at 50% of input.
OpenAI — GPT-5 Pro	`batchOutput`	Derived at 50% of output.
OpenAI — GPT-5 Nano	`cachedInput`	Derived at 10% of input.
OpenAI — GPT-5 Nano	`batchInput`	Derived at 50% of input.
OpenAI — GPT-5 Nano	`batchOutput`	Derived at 50% of output.
Google — Gemini 3 Flash	`cachedInput`	Derived at 10% of input — Google caching discount convention ~90%.
Google — Gemini 3.1 Flash-Lite	`cachedInput`	Derived at 10% of input — Google caching convention.
Google — Gemini 3.1 Flash-Lite	`batchInput`	Derived at 50% of input — Google Batch API uniform 50% discount.
Google — Gemini 3.1 Flash-Lite	`batchOutput`	Derived at 50% of output — Google Batch API uniform 50% discount.
Google — Gemini 2.5 Pro	`cachedInput`	Derived at 10% of input.
Google — Gemini 2.5 Flash	`cachedInput`	Derived at 10% of input.
Google — Gemini 2.5 Flash-Lite	`cachedInput`	Derived at 10% of input — Google caching convention.
Google — Gemini 2.5 Flash-Lite	`batchInput`	Derived at 50% of input — Google Batch API uniform 50% discount.
Google — Gemini 2.5 Flash-Lite	`batchOutput`	Derived at 50% of output — Google Batch API uniform 50% discount.
xAI — Grok 4 (legacy)	`cachedInput`	Extrapolated at 25% of base.

Vendor / Model

Field

Why it’s inferred

Anthropic — Claude Sonnet 4.6

cachedInput

Derived at 10% of input rate — Anthropic publishes 90% cache-hit discount on this tier.

Anthropic — Claude Sonnet 4.5

cachedInput

Derived at 10% of input rate; same 90% cache-hit convention as Sonnet 4.6.

Anthropic — Claude Sonnet 4.5

batchInput

Derived at 50% of standard input — Anthropic documents uniform 50% Batch discount.

Anthropic — Claude Sonnet 4.5

batchOutput

Derived at 50% of standard output — Anthropic documents uniform 50% Batch discount.

Anthropic — Claude Haiku 4.5

cachedInput

Derived at 10% of input rate — Anthropic 90% cache-hit discount convention.

OpenAI — GPT-5.4 Mini

cachedInput

Derived at 10% of input — OpenAI documents automatic 90% discount on cache hits across GPT-5.x tier.

OpenAI — GPT-5.4 Nano

cachedInput

Derived at 10% of input — OpenAI 90% cache-hit convention.

OpenAI — GPT-5.4 Nano

batchInput

Derived at 50% of input — OpenAI Batch API uniform 50% discount.

OpenAI — GPT-5.4 Nano

batchOutput

Derived at 50% of output — OpenAI Batch API uniform 50% discount.

OpenAI — GPT-5.4 Pro

cachedInput

Derived at 10% of input — OpenAI 90% cache-hit convention.

OpenAI — GPT-5.4 Pro

batchInput

Derived at 50% of input — OpenAI Batch API uniform 50% discount.

OpenAI — GPT-5.4 Pro

batchOutput

Derived at 50% of output — OpenAI Batch API uniform 50% discount.

OpenAI — GPT-5.2

cachedInput

Derived at 10% of input; no residency uplift.

OpenAI — GPT-5.2

batchInput

Derived at 50% of input.

OpenAI — GPT-5.2

batchOutput

Derived at 50% of output.

OpenAI — GPT-5

cachedInput

Derived at 10% of input.

OpenAI — GPT-5

batchInput

Derived at 50% of input.

OpenAI — GPT-5

batchOutput

Derived at 50% of output.

OpenAI — GPT-5.5 Pro

cachedInput

Derived at 10% of input — OpenAI does not publish a cached rate for *-pro models; using the family convention.

OpenAI — GPT-5.5 Pro

batchInput

Derived at 50% of input.

OpenAI — GPT-5.5 Pro

batchOutput

Derived at 50% of output.

OpenAI — GPT-5.2 Pro

cachedInput

Derived at 10% of input — pro-tier convention.

OpenAI — GPT-5.2 Pro

batchInput

Derived at 50% of input.

OpenAI — GPT-5.2 Pro

batchOutput

Derived at 50% of output.

OpenAI — GPT-5.1

batchInput

Derived at 50% of input.

OpenAI — GPT-5.1

batchOutput

Derived at 50% of output.

OpenAI — GPT-5 Pro

batchInput

Derived at 50% of input.

OpenAI — GPT-5 Pro

batchOutput

Derived at 50% of output.

OpenAI — GPT-5 Nano

cachedInput

Derived at 10% of input.

OpenAI — GPT-5 Nano

batchInput

Derived at 50% of input.

OpenAI — GPT-5 Nano

batchOutput

Derived at 50% of output.

Google — Gemini 3 Flash

cachedInput

Derived at 10% of input — Google caching discount convention ~90%.

Google — Gemini 3.1 Flash-Lite

cachedInput

Derived at 10% of input — Google caching convention.

Google — Gemini 3.1 Flash-Lite

batchInput

Derived at 50% of input — Google Batch API uniform 50% discount.

Google — Gemini 3.1 Flash-Lite

batchOutput

Derived at 50% of output — Google Batch API uniform 50% discount.

Google — Gemini 2.5 Pro

cachedInput

Derived at 10% of input.

Google — Gemini 2.5 Flash

cachedInput

Derived at 10% of input.

Google — Gemini 2.5 Flash-Lite

cachedInput

Derived at 10% of input — Google caching convention.

Google — Gemini 2.5 Flash-Lite

batchInput

Derived at 50% of input — Google Batch API uniform 50% discount.

Google — Gemini 2.5 Flash-Lite

batchOutput

Derived at 50% of output — Google Batch API uniform 50% discount.

xAI — Grok 4 (legacy)

cachedInput

Extrapolated at 25% of base.

AI cost and governance your SMB clients can afford

Is this you?

Plans

MSPCost AI Portfolio

Includes

What AICost changes in one AI workflow

AI front desk: phone, scheduling, intake and eligibility

Questions

Other paths

Builders and Small Teams

AI-Native Companies

Enterprises

System Integrators and Consultants

Talk to AICost

Methodology

Primary sources

Inferred values (marked with * in calculator tables)

Immediate steps you can take