Second-generation 30B model for Japanese enterprise text workloads, including document questions, extraction and summarization. Azure and private deployments have separate capability and billing contracts.
Azure guide page 29 calls max_tokens an output maximum while explicitly including prompt/system tokens. Preserve it as a shared request limit; independent input and output maxima remain unverified.
Maximum input tokens
Not collected
Maximum output
Not collected
32768 is documented as max_tokens including prompt/system tokens; not proven as an independent generated-output allowance.
Value in the provider’s env.sample, used by its streaming client. Managed endpoint is specific to the customer deployment.
Access
Managed online deployment via an Azure AI Hub project, requiring GPU quota. GPU uptime drives billing. Provider sample sets MODEL_NAME=tsuzumi2; exact deployed revisions are not pinned.
Regions
Japan-resident Azure tenants only. Guide warns about overseas data transfer without pinning a single serving region; choose and confirm the deployed region.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 14 Oct 2026
Category
Usage
Rate
Unit
Model surcharge
A100 GPU model surcharge1
$0.76
GPU hour
Model surcharge
H100 GPU model surcharge1
$0.76
GPU hour
Rate conditions
NTT DATA software/model surcharge for the specified GPU family, in USD. Azure GPU/VM infrastructure and other resources are additional. Billed while GPU is running, not per request or per token. A100 and H100 rows are alternatives, not two fees on one GPU. No-refund PayGo terms and region/subscription eligibility apply.
Guide says inference results are not used for other tenants’ models. No zero-retention claim verified. Marketplace excludes CSP, while the newer guide says availability can depend on the CSP provider; confirm eligibility.
Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms
Retention, training use, regional processing and enterprise terms have not been summarized for this endpoint.
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.
NTT DATA on Azure managed compute
tsuzumi2
Not verified
How it works
Not collected
Minimum reusable input
Not collected
Lifetime & refresh
Not collected
How to enable it
Not collected
Usage fields
Not collected
Billing rules
Not collected
Limits & data boundaries
Not collected
No verified cache rate for this endpoint. An empty rate is not a zero charge.
NTT group private deployment
Not verified
How it works
Not collected
Minimum reusable input
Not collected
Lifetime & refresh
Not collected
How to enable it
Not collected
Usage fields
Not collected
Billing rules
Not collected
Limits & data boundaries
Not collected
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
NTT identifies this as its proprietary, independently developed product model, distributed through commercial services and private deployments. No matching public checkpoint release was verified. Closed-weight classification is inferred for the exact product identity, not from provider ownership alone.
