Unverified50% confidenceFactExact time
Latxa is a family of Basque large language models released by the University of the Basque Country (UPV/EHU) and other institutions, built by continued pre-training and instruction fine-tuning of open-weight models such as Llama 2 on Basque corpora.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe model's backbone network uses Latxa, an existing Basque large language model, transferring text capability to the speech modality through fine-tuning.72% similarUnverifiedOllama and LM Studio are two tools that can run language models locally.67% similarUnverifiedResearch has already attempted to use LLMs to automatically translate natural language requirements into Alloy or TLA+ specification drafts, which are then reviewed and corrected by engineers.65% similarUnverifiedEarly text style transfer relied on GANs or VAEs and typically required parallel corpus training; whereas large language models such as the GPT series and Claude can complete high-quality tone conversion under zero-shot conditions through instruction fine-tuning and RLHF65% similarUnverifiedMainstream international open-source models such as Qwen, Gemma, and Llama 3, whose training data includes multilingual data, can meet basic application requirements on Japanese tasks.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1106302API
curl https://kongchang.com/api/v1/knowledge/claims/1106302MCP
get_claim(id=1106302)