{"id": "na-model-deepseek-v32", "key": "neuroai2026modeldeepseekv32", "title": "DeepSeek V3.2 bets on sparse attention for cheaper long context", "author": "NeuroAI", "publisher": "BrainNet (脑机网)", "date": "2026-10-24", "url": "https://neuroai.site/a/na-model-deepseek-v32", "section": "Foundation Models", "apa": "NeuroAI. (2026, October 24). DeepSeek V3.2 bets on sparse attention for cheaper long context. BrainNet (脑机网). https://neuroai.site/a/na-model-deepseek-v32", "mla": "\"DeepSeek V3.2 bets on sparse attention for cheaper long context.\" NeuroAI, BrainNet (脑机网), 24 Oct. 2026, neuroai.site/a/na-model-deepseek-v32.", "chicago": "\"DeepSeek V3.2 bets on sparse attention for cheaper long context.\" NeuroAI, BrainNet (脑机网), October 24, 2026. https://neuroai.site/a/na-model-deepseek-v32.", "plain": "NeuroAI (脑机网), \"DeepSeek V3.2 bets on sparse attention for cheaper long context,\" 24 October 2026, https://neuroai.site/a/na-model-deepseek-v32", "bibtex": "@misc{neuroai2026modeldeepseekv32,\n  author       = {{NeuroAI}},\n  title        = {{DeepSeek V3.2 bets on sparse attention for cheaper long context}},\n  year         = {2026},\n  month        = {oct},\n  publisher    = {{BrainNet (脑机网)}},\n  howpublished = {\\url{https://neuroai.site/a/na-model-deepseek-v32}},\n  note         = {China AI Decoded. Accessed 2026-10-24.},\n  language     = {english}\n}\n", "ris": "TY  - GEN\nTI  - DeepSeek V3.2 bets on sparse attention for cheaper long context\nAU  - NeuroAI\nPY  - 2026\nDA  - 2026/10/24\nPB  - BrainNet (脑机网)\nUR  - https://neuroai.site/a/na-model-deepseek-v32\nLA  - en\nER  - \n", "exports": {"bibtex": "https://neuroai.site/a/na-model-deepseek-v32.bib", "ris": "https://neuroai.site/a/na-model-deepseek-v32.ris", "markdown": "https://neuroai.site/a/na-model-deepseek-v32.md", "json": "/api/cite/na-model-deepseek-v32"}}