Back to articles
DevPickView feed
gihyo.jp
Original published:

PFN releases model weights for translation-specialized LLM 'PLaMo 3 Translate', available in 4 sizes from 0.8B to 31B

PFN releases model weights for translation-specialized LLM 'PLaMo 3 Translate', available in 4 sizes from 0.8B to 31B
AI summary
Editorial policy

PFN released the model weights for its translation-specialized LLM 'PLaMo 3 Translate' on Hugging Face, offering four sizes from 0.8B to 31B for local execution.

Key points

  • •Preferred Networks (PFN) released the model weights for 'PLaMo 3 Translate', a translation-specialized LLM, on Hugging Face on November 5, 2026.
  • •The model is available in four sizes: 0.8B, 2B, 8B, and 31B (where B stands for billion parameters).
  • •Users can select a model according to their environment, download it, and run it locally.
Read the original(opens in a new tab)
#LLM / Model Trends#Major Players#PLaMo 3 Translate#Preferred Networks

Related articles

See more
ITmedia AI+

PFN Open-Sources Translation LLM Claiming "World-Class" Performance, 4 Sizes from 0.8 to 31B, On-Device Compatible

Preferred Networks open-sourced the translation-specialized LLM 'PLaMo 3 Translate' with open weights, claiming world-class performance.

ITmedia AI+

PFN announces domestic flagship model 'PLaMo 3.1 Prime' with on-premises Deep Research support and free API plan

Preferred Networks began offering its domestic generative AI foundation model 'PLaMo 3.1 Prime', featuring on-premises Deep Research capabilities and a new free API plan.

ITmedia AI+

Expanding use cases of 'domestic AI': The 'pitfalls' seen in PFN's 'PLaMo'

The article explains use cases and practical challenges of Preferred Networks' domestic AI 'PLaMo', based on the company's exhibition at CEATEC 2026.

PC Watch

Japanese-Specialized Domestic Translation AI 'PLaMo Translation' Achieves State-of-the-Art Performance with Update

Preferred Networks significantly updated its Japanese-focused translation service, PLaMo Translation, achieving accuracy comparable to state-of-the-art models.

INTERNET Watch

Akamai Launches AI Inference Service 'Akamai Inference' and AI Agent Execution Environment 'Akamai Agent Sandbox'

Akamai announced new products providing an AI inference environment and an isolated execution environment for agents on its cloud service, supporting enterprises from model building to operation.