Skip to content
Model infoModelsSnapshot: 15 Oct 2026

MiniMax / M2-her

MiniMax M2-her

Dialogue-specialized model with role settings and multi-turn history. The current model FAQ lists it as retired.

Language models (LLMs)
retiredClosed weightsText generation
Official documentation (opens in a new tab)Reviewed 12 Oct 2026
3 of 16 specification fields collected
Context window64K64,000 tokens
Maximum outputNot collectedAwaiting source review
Input billingNot collectedMiniMax
Output billingNot collectedNo inferred price

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
64,000 tokensSource for Context / input limit (opens in a new tab)
Maximum output
Not collected
Input modalities
TextSource for Input modalities (opens in a new tab)
Output modalities
TextSource for Output modalities (opens in a new tab)
Reasoning
Not collected
Tool calling
Not collected
Structured output
Not collected
Streaming
Not collected
Batch processing
Not collected
Fine-tuning
Not collected
Knowledge cutoff
Not collected
Release date
Not collected
Parameters
Not collected
Architecture
Not collected
Languages
Not collected
Context caching
Not collected

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

MiniMax

Provider source (opens in a new tab)
API model ID
M2-her
Access
Retired according to the current model FAQ. Guide retained as historical documentation.
Regions
Not collected
API endpoint URL
POST https://api.minimax.io/v1/chat/completionsSource for API endpoint URL (opens in a new tab)

Endpoint pricing has not been collected. Check the provider for applicable rates.

Data & enterprise terms

The March 30, 2026 service terms retain provider IP and allow use of inputs and generated content to provide, maintain, develop and improve the service. This review does not establish a no-training or zero-retention guarantee.

Read the service policy (opens in a new tab)

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

MiniMax

M2-her
Not verified

No model-specific cache contract verified for this endpoint.

How it works
Not collected

No model-specific cache contract verified for this endpoint.

Minimum reusable input
Not collected

No model-specific cache contract verified for this endpoint.

Lifetime & refresh
Not collected

No model-specific cache contract verified for this endpoint.

How to enable it
Not collected

No model-specific cache contract verified for this endpoint.

Usage fields
Not collected

No model-specific cache contract verified for this endpoint.

Billing rules
Not collected

No model-specific cache contract verified for this endpoint.

Limits & data boundaries
Not collected

No model-specific cache contract verified for this endpoint.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

Inferred closed-weight service variant: the exact model is documented through MiniMax service access and no corresponding weight release was identified in the reviewed official MiniMaxAI repository inventory. Service terms reserve underlying IP. Public M-series, H3 and Music 3 weights are explicitly excluded; the OpenAPI MIT label licenses documentation, not model weights.

Classification evidence (opens in a new tab)

Sources reviewed