Skip to content
Model infoModelsSnapshot: 15 Oct 2026

ABEJA / ABEJA GPT

ABEJA GPT 13B

Original Japanese autoregressive GPT model from ABEJA’s 2022 development program. Current production access to this exact generation is unverified.

Language models (LLMs)
historicalClosed weightsText generation
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
8 of 20 specification fields collected
Context windowNot collectedAwaiting source review
Maximum outputNot collectedAwaiting source review
Input billingNot collectedABEJA original GPT program (historical)
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
Not collected

The architecture article gives 2048-token settings for small and medium experiments, not a verified 13B serving limit.

Maximum input tokens
Not collected
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
2022-07Source for Release date (opens in a new tab)

Public announcement month; the technical report was published July 27 after ABEJA SIX 2022.

Parameters
13BSource for Parameters (opens in a new tab)
Architecture
GPT-NeoX-based transformer with selected bias removal, tied input/output embeddings and attention dropout 0.1Source for Architecture (opens in a new tab)

Final 13B design described in section 7. SwishGLU and parallel transformer layers were tested but not adopted. Dropout is a training setting.

Languages
JapaneseSource for Languages (opens in a new tab)

Documented focus, not an exhaustive language guarantee.

Context caching
Not collected

No model-specific prompt-caching contract or cached-input price verified. Training caches and recurrent-state dtype settings are not a published prompt-cache API.

Published price
Not collected

No numeric production price verified for this exact model or research checkpoint.

providerModelLabel
ABEJA model4Source for providerModelLabel (opens in a new tab)

Original comparison table uses this label; the Atlas display name adds the published size for clarity.

Published training data
Japanese Wikipedia, CC100 and OSCAR; mC4 used in some program versionsSource for Published training data (opens in a new tab)

Provider describes the program-level sources but does not pin every mixture or token count to this exact size.

Access & pricing

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

ABEJA original GPT program (historical)

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

Provider display label is not a verified API request selector.

Access
Provider reported internal use and invited business discussions. No live self-service API, current exact-checkpoint availability or public price is verified.
Regions
No production processing region or deployment agreement verified for this exact checkpoint.

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.

ABEJA original GPT program (historical)

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

Provider identifies this model or selected research checkpoint, but no matching public weights were found in the reviewed official inventory. Closed-weight classification is inferred for this exact artifact. Public base weights, training recipes and other ABEJA releases do not establish public availability of this derivative; private distribution terms remain unverified.

Classification evidence (opens in a new tab)

Lifecycle dates

Original development report published27 Jul 2022Source (opens in a new tab)
Final 13B architecture report published10 Aug 2022Source (opens in a new tab)

Sources reviewed