Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Preferred Networks / PLaMo

PLaMo 3.0 Prime

Current Prime generation for Japanese text, reasoning and agent workflows, available through the PLaMo API and private deployments.

Language models (LLMs)
availableClosed weightsText generation
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
12 of 19 specification fields collected
Context window262.1K262,144 tokens
Maximum output20.5K20,480 tokens
Input billingSee ratesPLaMo API
Output billingSee ratesPublished units and conditions below

Specifications

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

Context / input limit
262,144 tokensSource for Context / input limit (opens in a new tab)

Shared input plus requested output budget.

Maximum input tokens
Not collected
Maximum output
20,480 tokensSource for Maximum output (opens in a new tab)
Input modalities
TextSource for Input modalities (opens in a new tab)
Output modalities
TextSource for Output modalities (opens in a new tab)
Reasoning
SupportedSource for Reasoning (opens in a new tab)

none or medium; default none.

Tool calling
SupportedSource for Tool calling (opens in a new tab)
Structured output
SupportedSource for Structured output (opens in a new tab)

JSON Schema with strict mode; supported schema subset applies.

Streaming
SupportedSource for Streaming (opens in a new tab)

Server-sent events.

Batch processing
Not collected
Fine-tuning
Not collected
Knowledge cutoff
Not collected
Release date
2026-06-22Source for Release date (opens in a new tab)
Parameters
Not collected
Architecture
Not collected
Languages
Japanese, EnglishSource for Languages (opens in a new tab)

Languages explicitly evaluated, not an exhaustive supported-language list.

Context caching
Not collected

No model-specific public prompt-cache contract or cached-token discount was verified. Internal inference caches do not establish a customer-facing cache feature.

parallelToolCalling
Not supportedSource for parallelToolCalling (opens in a new tab)

Provider excludes parallel-call categories from its BFCL evaluation because parallel calls are unsupported.

Published price
JPY 60 input / 250 output per million tokensSource for Published price (opens in a new tab)

128k price band; promotional parity above it. Free plan not yet available.

Access & pricing

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

PLaMo API

Provider source (opens in a new tab)
API model ID
plamo-3.0-prime
Access
Standard paid plan or sales-approved Provider plan. The listed Free plan is still in preparation.
Regions
Standard registration requires Japanese residence or a Japanese legal entity; use outside Japan needs provider permission under the terms.
apiBase
https://api.platform.preferredai.jp/v1Source for apiBase (opens in a new tab)
rateLimit
100 requests per minute per API keySource for rateLimit (opens in a new tab)

Published baseline; plan and approved adjustments can differ.

reasoningBilling
Reasoning-summary generation also consumes tokensSource for reasoningBilling (opens in a new tab)
Creation privacy
Standard: input not retained or used for trainingSource for Creation privacy (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Standard text APIInput1¥601M tokens
Standard text APIOutput1¥2501M tokens

Rate conditions

  1. Published up to 128k tokens. Above 128k the same price applies during the release promotion; post-promotion rates and end date are not established. Provider rounds fractional yen upward when billing; no per-request rounding rule is inferred.
Pricing source (opens in a new tab)

Data & enterprise terms

The Standard pricing page promises no storage or training use of submitted input. Eligibility and permitted use are governed by the PLaMo 3 terms; these conditions do not establish every private-host policy.

Read the service policy (opens in a new tab)

Microsoft Foundry

Provider source (opens in a new tab)
API model ID
Not collected

Display or catalog name is not a verified API request selector.

Access
Provider catalog version 1; private GPU deployment. Token prices from the PLaMo API do not apply here.
Regions
Deployment and service contract determine access and processing location.
catalogModel
plamo-30-prime / version 1 / azureml-preferrednetworks-pSource for catalogModel (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Model license surchargeNVIDIA A1001$3.99A100 GPU hour
Model license surchargeNVIDIA H1001$6.99H100 GPU hour

Rate conditions

  1. Per GPU, plus Azure infrastructure costs. Not an all-in GPU price or a token tariff.
Pricing source (opens in a new tab)

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.

PLaMo API

plamo-3.0-prime
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.

Caching reviewed 14 Oct 2026

Microsoft Foundry

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.

Caching reviewed 14 Oct 2026

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

Named provider model offered through commercial services, private licensing or provider research, with no matching public checkpoint verified. Closed-weight classification is inferred for this specific derivative, not its public base or other PLaMo releases.

Classification evidence (opens in a new tab)

Lifecycle dates

Announced22 Jun 2026Source (opens in a new tab)
Microsoft Foundry offering announced6 Oct 2026Source (opens in a new tab)

Sources reviewed