Original Japanese enterprise model co-developed with Cohere and launched through Fujitsu Kozuchi and DI PaaS. The launch does not pin a numbered checkpoint.
No numeric per-token price verified for this exact model. Enterprise service quotes and limited platform trials do not establish free production inference.
The published display name is not a verified API request selector. Tenant endpoint URLs and keys are issued by the provider.
Access
Historical launch offering for secure private enterprise deployments; exact original revision availability must be confirmed.
Regions
Service, contract and deployment determine location and access.
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.
Fujitsu Kozuchi / DI PaaS
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
JSTS
Fujitsu · vendor-reported
0.93Pearson correlation
September 2024 launch table, as published by Fujitsu. Model snapshot and complete evaluation configuration are not pinned. Scores are proportions, not percentages; the provider labels JSQuAD as Accuracy. These are not independently reproduced or assigned to later Takane revisions.
September 2024 launch table, as published by Fujitsu. Model snapshot and complete evaluation configuration are not pinned. Scores are proportions, not percentages; the provider labels JSQuAD as Accuracy. These are not independently reproduced or assigned to later Takane revisions.
September 2024 launch table, as published by Fujitsu. Model snapshot and complete evaluation configuration are not pinned. Scores are proportions, not percentages; the provider labels JSQuAD as Accuracy. These are not independently reproduced or assigned to later Takane revisions.
September 2024 launch table, as published by Fujitsu. Model snapshot and complete evaluation configuration are not pinned. Scores are proportions, not percentages; the provider labels JSQuAD as Accuracy. These are not independently reproduced or assigned to later Takane revisions.
September 2024 launch table, as published by Fujitsu. Model snapshot and complete evaluation configuration are not pinned. Scores are proportions, not percentages; the provider labels JSQuAD as Accuracy. These are not independently reproduced or assigned to later Takane revisions.
September 2024 launch table, as published by Fujitsu. Model snapshot and complete evaluation configuration are not pinned. Scores are proportions, not percentages; the provider labels JSQuAD as Accuracy. These are not independently reproduced or assigned to later Takane revisions.
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
Takane is a Fujitsu-specific enterprise model developed with Cohere, offered through private services and deployments. No matching public Takane checkpoint release was verified. Classification is inferred for the named Takane product, not for public Command-family weights or every Fujitsu research model.
