---
title: "MiniMax Agent | Ethical Stack"
url: https://ethicalstack.ai/products/minimax-agent/
catalog_version: f6baff2119d0e9158520da7fbe4232011dac79f25bd08757436d7a1dbd593db2
---

# MiniMax Agent

Hosted general-purpose agent for research, document and creative tasks.

Canonical profile: https://ethicalstack.ai/products/minimax-agent/

Scope: MiniMax Agent web and desktop application; excludes Open Platform APIs and separate MiniMax Code tools.

Published overall: 3.4; 5/6 dimensions rated.

Round the equal-weight mean of available numerical ratings to one decimal; require at least four of six for an overall score. BDS status stays outside the mean; for eligible profiles, a boycott target caps the final score at 2/10 and a pressure target subtracts one point (minimum 1). Not listed and not checked do not adjust the score.

## Energy practices

Finding: 3/10

Assessment: https://ethicalstack.ai/products/minimax-agent/#energy

Public docs establish the MiniMax Agent web and desktop service route; they do not establish its request-level physical footprint. The product page identifies the selected model and exposes task/workflow choices; these configure use, but no energy reduction is measured.

Reason: 1 + 1 reporting + 0 electricity + 1 controls = 3/10. This grades documented practices and transparency.

Argument: This practices and disclosure grade covers MiniMax Agent web and desktop service. Public product/API documentation establishes how this route operates and the listed user controls, but gives no measured per-request energy, emissions or water. A corporate electricity claim is not allocated to this route.

Type: argument; confidence: medium

Scope (dimension): MiniMax Agent web and desktop service

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Request-level electricity, emissions, water use, serving utilization and any route-specific clean-power matching are not reported for MiniMax Agent web and desktop service.

```json
{
  "method": "practices-transparency-v1",
  "scope": "MiniMax Agent web and desktop service",
  "components": {
    "disclosure": {
      "score": 1,
      "reason": "Official documentation describes the MiniMax Agent web and desktop service operating boundary, but does not publish a dated quantitative energy, emissions or water result for this route.",
      "sourceUrls": [
        "https://agent.minimax.io/"
      ]
    },
    "electricity": {
      "score": 0,
      "reason": "The reviewed sources do not map MiniMax Agent web and desktop service to a site-specific electricity mix or time-matched clean-power supply.",
      "sourceUrls": [
        "https://agent.minimax.io/"
      ]
    },
    "demand": {
      "score": 1,
      "reason": "The product page identifies the selected model and exposes task/workflow choices; these configure use, but no energy reduction is measured.",
      "sourceUrls": [
        "https://agent.minimax.io/"
      ]
    }
  },
  "actualImpactMeasured": false
}
```

- [MiniMax Agent product page](https://agent.minimax.io/) (primary; checked 2026-10-24)

## Training data

Finding: Depends on the model

Assessment: https://ethicalstack.ai/products/minimax-agent/#training-rights

Agent model versions may change; the app policy does not describe training corpus rights for each selected model.

Reason: Agent model versions may change; the app policy does not describe training corpus rights for each selected model.

Argument: The MiniMax app privacy policy describes information collected from users and feature operation, but it does not provide corpus provenance or permissions for the model served by MiniMax Agent. Submitted user data and original model training data are separate scopes.

Type: argument; confidence: medium

Scope (profile): MiniMax Agent web and desktop application; excludes Open Platform APIs and separate MiniMax Code tools.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Exact underlying model version, pretraining sources, source permissions, post-training datasets, and creator remedies are not established in the app sources.

```json
{
  "method": "training-rights-evidence-v1",
  "scope": "The stated training release or selected model",
  "components": null,
  "routeDependency": "Agent model versions may change; the app policy does not describe training corpus rights for each selected model.",
  "modelReference": null
}
```

- [MiniMax AI App and Web Privacy Policy](https://agent.minimax.io/doc/en/privacy-policy.html) (primary; checked 2026-10-24)

- [MiniMax Agent product page](https://agent.minimax.io/) (primary; checked 2026-10-24)

## Your data & privacy

Finding: 4/10

Assessment: https://ethicalstack.ai/products/minimax-agent/#privacy

The app policy covers web, mobile and desktop services and describes account, usage, IP, logs, and user-provided data. Agent features may access local files and connected browser/account data when enabled.

Reason: The app terms disclose meaningful collection and agent access boundaries, but prompt retention and route-specific model handling are not fixed.

Argument: MiniMax’s app policy applies to its web and downloaded applications, collects account and usage data, and says features may access local data and connected applications when enabled. The Agent terms describe browser operations and a pause before specified sensitive actions. Retention remains purpose-based rather than a fixed prompt period.

Type: argument; confidence: medium

Scope (profile): MiniMax Agent web and desktop application; excludes Open Platform APIs and separate MiniMax Code tools.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: The policy does not give a fixed retention period for each Agent payload or connected-account record; local-access settings and feature-specific changes should be reviewed.

- [MiniMax AI App and Web Privacy Policy](https://agent.minimax.io/doc/en/privacy-policy.html) (primary; checked 2026-10-24)

- [MiniMax AI App and Web Terms of Service](https://agent.minimax.io/doc/en/terms-of-service.html) (primary; checked 2026-10-24)

## Openness

Finding: 2/10

Assessment: https://ethicalstack.ai/products/minimax-agent/#openness

A hosted agent product with no self-hostable service runtime in the reviewed product materials.

Reason: Hosted agent interface without self-hostable app or service runtime.

Argument: The consumer app provides agent capabilities through MiniMax-hosted web and desktop interfaces. The reviewed product pages do not expose a deployable app or serving runtime; separate model weights and APIs have their own records.

Type: argument; confidence: medium

Scope (profile): MiniMax Agent web and desktop application; excludes Open Platform APIs and separate MiniMax Code tools.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: MiniMax model releases and Open Platform endpoints are distinct from the Agent app.

- [MiniMax Agent product page](https://agent.minimax.io/) (primary; checked 2026-10-24)

- [MiniMax AI App and Web Privacy Policy](https://agent.minimax.io/doc/en/privacy-policy.html) (primary; checked 2026-10-24)

- [MiniMax AI App and Web Terms of Service](https://agent.minimax.io/doc/en/terms-of-service.html) (primary; checked 2026-10-24)

## Worker protections

Finding: 1/10

Assessment: https://ethicalstack.ai/products/minimax-agent/#labour

MiniMax’s reviewed company pages and M3.1 Flash model license govern company/product or user matters, not employee or supplier labor conditions.

Reason: 1 + 0 pay & conditions + 0 worker voice & remedy + 0 evidence & follow-through = 1/10. This grades documented worker safeguards, not a measured rate of mistreatment.

Argument: The company’s consumer terms and its M3.1 Flash community license govern customer or model use, not employment. No current corporate ESG report, worker grievance policy, supplier labor standard, or worker-outcome disclosure was established in the checked company material. The score therefore records a narrow public-disclosure gap and makes no claim about actual treatment. This entry carries the same developer/operator finding to MiniMax Agent; it does not establish that this product was deployed under every cited relationship.

Type: argument; confidence: low

Scope (dimension): Nanonoble Pte. Ltd. (MiniMax): the documented developer/operator record, not proof of deployment of MiniMax Agent.

Research reviewed: 2026-10-24; evidence date: not recorded

Unknowns: Employee and contractor conditions, data annotation/evaluation labor, workforce voice, grievance procedures, supplier coverage, and labor audits remain unresolved.

```json
{
  "method": "worker-safeguards-evidence-v1",
  "entity": "minimax",
  "company": "MiniMax",
  "scope": "MiniMax's workforce and disclosed contractors. Coverage of individual product teams may be incomplete.",
  "components": {
    "conditions": {
      "score": 0,
      "reason": "No worker-condition or supply-chain labor disclosure was established in the reviewed company materials; consumer/model terms are excluded from this component.",
      "sourceUrls": [
        "https://www.minimax.io/terms-of-service-v2.html",
        "https://www.minimax.io/"
      ]
    },
    "voice": {
      "score": 0,
      "reason": "The reviewed company pages do not disclose a worker grievance or representation mechanism; the matched M3.1 model license is not worker evidence.",
      "sourceUrls": [
        "https://www.minimax.io/terms-of-service-v2.html",
        "https://www.minimax.io/"
      ]
    },
    "assurance": {
      "score": 0,
      "reason": "No labor-specific audit, worker-level corroboration, or remediation outcome was established in the reviewed material.",
      "sourceUrls": [
        "https://www.minimax.io/terms-of-service-v2.html",
        "https://www.minimax.io/"
      ]
    }
  },
  "disclosureLimited": true,
  "meaning": "This grades documented worker safeguards, not a measured rate of mistreatment."
}
```

- [MiniMax consumer Terms of Service](https://www.minimax.io/terms-of-service-v2.html) (primary; checked 2026-10-24)

- [MiniMax official company website](https://www.minimax.io/) (primary; checked 2026-10-24)

## Military ties

Finding: 7/10

Assessment: https://ethicalstack.ai/products/minimax-agent/#military-contracts

The Agent terms require lawful use and allow enforcement for violations, but do not establish a military-purpose ban, military-specific oversight, or a disclosed military contract.

Reason: 7 base + 0 safeguard adjustment = 7/10 (0 of 9 safeguard points; every 3 adds 1). No military relationship was established in the cited sources. This ranks above known defense ties without claiming a verified clean record.

Argument: The MiniMax AI App and Web Terms govern the Agent service and prohibit unlawful use, but do not identify military use as a prohibited purpose or disclose a military customer, military-specific monitoring, or enforcement record. The M3.1 Flash repository license is not applied to this separate app route.

Type: argument; confidence: low

Scope (dimension): MiniMax Agent app route under the AI App and Web Terms; separate from M3.1 Flash model license and Open Platform API.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Customer contracts and downstream uses may not be public. The sources reviewed do not establish whether there are military customers, deployments, or service-specific controls beyond the published general terms.

```json
{
  "method": "military-ties-priority-v2",
  "entity": "minimax",
  "company": "MiniMax",
  "scope": "MiniMax Agent app route under the AI App and Web Terms; separate from M3.1 Flash model license and Open Platform API.",
  "components": {
    "restrictions": {
      "score": 0,
      "reason": "The route terms require lawful or properly qualified use but do not state a military-purpose prohibition. The M3.1 Flash repository license is model-specific and is not applied to this hosted route.",
      "sourceUrls": [
        "https://agent.minimax.io/doc/en/terms-of-service.html"
      ]
    },
    "oversight": {
      "score": 0,
      "reason": "The reviewed terms allow general enforcement or suspension but disclose no military-specific screening, monitoring, or remedy for this route.",
      "sourceUrls": [
        "https://agent.minimax.io/doc/en/terms-of-service.html"
      ]
    },
    "transparency": {
      "score": 0,
      "reason": "The reviewed route terms do not identify a military customer, contract, or deployment; this does not establish that none exists.",
      "sourceUrls": [
        "https://agent.minimax.io/doc/en/terms-of-service.html"
      ]
    }
  },
  "disclosureLimited": true,
  "relationship": "not-established",
  "relationshipReason": "The reviewed MiniMax Agent or Open Platform route terms do not establish a direct military customer, contract, or deployment. This is limited to the published route terms and does not assert none exists.",
  "score": 7,
  "baseScore": 7,
  "safeguardPoints": 0,
  "safeguardAdjustment": 0,
  "rawScore": 7,
  "cap": null
}
```

- [MiniMax AI App and Web Terms of Service](https://agent.minimax.io/doc/en/terms-of-service.html) (primary; checked 2026-10-24)

## BDS targets

Finding: Not listed

Assessment: https://ethicalstack.ai/products/minimax-agent/#israel-palestine

We did not find MiniMax among the named targets in the two official BNC campaign lists checked on 2026-10-23. This is a limited list check, not BDS approval.

Reason: BDS campaign status is a factual category, not a 1–10 ethical rating.

Argument: MiniMax is not a named target in the checked lists, but BDS uses selective campaigns, so absence cannot establish a clean record. Running downloaded weights locally is a different decision from paying the developer or a cloud host; check the host as well. This entry carries the same developer/operator finding to MiniMax Agent; it does not establish that this product was deployed under every cited relationship.

Type: evidence; confidence: medium

Scope (dimension): Nanonoble Pte. Ltd. (MiniMax): the documented developer/operator record, not proof of deployment of MiniMax Agent.

Research reviewed: 2026-10-23; evidence date: not recorded

Unknowns: This review covers the named targets in the BNC consumer/pressure guide and No Tech campaign, not every BDS campaign, local group, investment or supplier. Lists and ownership can change. A company’s absence from these lists is not an endorsement or proof that it has no relevant relationships.

- [BNC target list (not checked for this company)](https://bdsmovement.net/Guide-to-BDS-Boycott) (campaign; checked 2026-10-24)

- [BNC target list (not checked for this company)](https://bdsmovement.net/no-tech-oppression-apartheid-or-genocide) (campaign; checked 2026-10-24)

## Reading the evidence

- Assessments are provisional editorial judgments. Cite the profile and the underlying sources, and retain confidence, scope and unknowns.

- Unknown, not-applicable and model-dependent are different states. Missing evidence does not establish good or bad conduct.

- Company-level evidence does not establish use of a particular product or model. Account plan, model, host and route can change an assessment.

- Energy scores describe documented practices and transparency, not measured request emissions. Zero component points can mean an evidence gap.

- Published overall scores require at least four of six numerical dimensions. Preserve coverage and BDS adjustments; BDS itself is categorical.

- No named BDS target found is limited to the reviewed campaign sources and is not approval. Personalized priorities do not change the published overall score.

- Review and source-check dates describe the research, not this deployment. Fetch current primary sources when the decision requires current verification.

Methodology: https://ethicalstack.ai/methodology/