# Agentforce vs Level AI: who uses which

On usedby, 41 companies use Agentforce and 20 use Level AI. 1 company uses both, including OpenTable. Counts include every proof level, job-post signals included; each company keeps its proof level on each side. Latest check 2026-11-16: 61 of 61 public lines checked.

## Side by side

|  | [Agentforce](https://www.usedby.ai/tools/agentforce.md) | [Level AI](https://www.usedby.ai/tools/level-ai.md) |
|---|---|---|
| Public customers | 41 | 20 |
| Confirmed | 32 | 13 |
| Observed | 8 | 7 |
| Signals (job posts) | 1 | 0 |
| Top industries of customers (every proof level) | Consumer Goods (4), Insurance & InsurTech (3), Accounting & Audit (2) | Fintech & Payments (4), E-commerce & D2C (3), HR & Recruiting (2) |
| Category | AI agents for customer service | Contact center AI |
| Data processing terms | A data processing agreement is available. ([salesforce.com](https://www.salesforce.com/company/legal/privacy/?bc=OTH), checked 2026-11-15) | A data processing agreement is available. ([thelevel.ai](https://thelevel.ai/legal/dpa), checked 2026-11-02) |
| Free trial | Free trial or free plan available. ([salesforce.com](https://www.salesforce.com/agentforce/pricing/?bc=OTH), checked 2026-11-15) | — |
| Pricing | Flex Credits: $ 500 USD per 100,000 credits ([salesforce.com](https://www.salesforce.com/agentforce/pricing/), checked 2026-11-16) | — |
| Your data | — | Not used to train models by default. ([thelevel.ai](https://thelevel.ai/security), checked 2026-11-02) |
| Integrations | — | OneLogin, Azure, Okta, Guru, Zoom, Auth0, Lessonly, Slack, Intercom, Five9, Twilio, Amazon Connect ([thelevel.ai](https://thelevel.ai/integrations), checked 2026-11-02) |

## Companies that use both

Every proof level counts; the proof level of each side is shown, strongest proofs first.

| Company | Proof, Agentforce | Proof, Level AI |
|---|---|---|
| [OpenTable](https://www.usedby.ai/companies/opentable.md) | Confirmed | Observed |

## Published results

Reviewed stories only; usedby did not measure the figures.

- Agentforce: [Salesforce](https://www.usedby.ai/case-studies/salesforce-agentforce.md): **$20 million**: “$20 million lift to marketing-driven sales pipeline” (captured by usedby 2026-11-15)
- Agentforce: [Greenlane](https://www.usedby.ai/case-studies/greenlane-agentforce.md): **58%**: “Agentforce delivered answers 58% faster than Greenlane's shared call center model” (captured by usedby 2026-11-15)
- Level AI: [ezCater](https://www.usedby.ai/case-studies/ezcater-level-ai.md): **13%**: “ezCater has seen a 13% decrease in its overall call handling time” (captured by usedby 2026-11-16)
- Level AI: [Sungage Financial](https://www.usedby.ai/case-studies/sungage-financial-level-ai.md): **100%**: “100% automated QA coverage” (captured by usedby 2026-11-16)

## Evidence for your decision: Level AI

Source observations. Check the required plan and conditions before choosing.

### Documented tasks

- Topcon — Resolve customer support requests
  > Instead of manually programming answers, the team integrated Level AI’s Virtual Agent directly with Topcon's massive library of 12,000 to 15,000 knowledge assets.
  [Source](<https://thelevel.ai/case-studies/topcon-positioning-system>) · VENDOR CASE STUDY · Captured: 2026-10-08T01:03:03.219Z

### Choice criteria

- Integration: Salesforce — Sourced observation
  Plan: Unknown · Checked: 2026-10-24T09:13:06.554933+00:00
  > telephony systems and CRM (Salesforce)
  [Source](<https://thelevel.ai/case-studies/topcon-positioning-system>) · Documented · Not tested
  Product: Level AI Topcon deployment
  Setup and limits: Documented only; not\_tested. No authenticated vendor operation executed.; No capability profile or execution evidence inferred.
  Conditions: Documentary statement only; named account permissions and regional eligibility unknown.; Case deployment scope only; general connector availability, exact operations and plan entitlement unknown.; Product scope needs verification


## Capabilities for agents: Level AI

Documentation and operational tests are separate. A test covers only the stated task and operation.

No reviewed capability profile yet.

---
Source: https://www.usedby.ai/compare/agentforce-vs-level-ai · How we check: https://www.usedby.ai/methodology
