Skip to content
← Models
Open weightsStability AI/10 trustMetrics-checked

Japanese StableLM 7B

Stability AI's 7B Japanese-first LLM, Apache 2.0 for the base model.

Visit official page →

About

Japanese StableLM Alpha was released by Stability AI on August 10, 2023, the company's first non-English language model. The 7-billion-parameter model was trained on 750 billion tokens of Japanese and English web text using an extended GPT-NeoX stack with SwiGLU and xPos, plus datasets built with Stability AI Japan's community and the EleutherAI Polyglot Japanese team. Stability AI reported it as the top-scoring publicly available Japanese LM across eight lm-evaluation-harness tasks.

Two variants shipped: Japanese StableLM Base Alpha 7B under Apache 2.0 for commercial use, and an Instruct Alpha 7B for research only. Follow-ups included Japanese Stable LM 3B-4E1T and Gamma 7B.

Specs

Context window

Pricing

Free quota

API access

Weights

Open weights

Released

August 10, 2023

Benchmarks

Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.

Use cases

japanese-nlptext-generationresearchcommercial-usechatbots

Trust score — 7/10

/10 trustMetrics-checked

Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 119; Trending score 68/100 from researcher assessment.

Sources