MiniMax M3, an open-weight LLM from MiniMaxAI, has been released with a context window of over one million tokens. Priced at $0.30 per million input tokens, this model aims to make it economically feasible to process large amounts of data, such as entire codebases, directly within the prompt. Initial tests indicate strong performance in coding tasks, arithmetic reasoning, and strict JSON extraction, suggesting its utility for applications like document analysis and code review where large context processing is beneficial. AI
IMPACT Enables cost-effective processing of large documents and codebases directly in prompts, potentially reducing reliance on RAG pipelines for certain tasks.
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 →