Tencent has launched its Hy4 flash large language model, featuring 224 billion total parameters and 12 billion active parameters. The model is priced at 2 yuan per million output tokens and demonstrates strong performance on Tencent's internal WorkIndex benchmark. AI
IMPACT This release expands the parameter count and performance benchmarks for large language models, particularly within the Chinese market.
RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →