DeepSeek is internally beta testing its new DeepSeek V4.1 Flash model, which features a novel architecture with native multimodal support. This intermediate version promises enhanced capabilities, increased speed, and reduced costs compared to previous models. Users can access the API with a specific model name, and current pricing matches that of DeepSeek V4 Flash, with a rate limit of 20 concurrent requests per account. AI
IMPACT This new model release from DeepSeek could signal advancements in multimodal AI capabilities and cost-efficiency.
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 →