DeepSeek-V4-Pro-Max beats GPT-5.2 and Gemini 3.0-Pro on standard reasoning benchmarks and lands just behind GPT-5.4 and Gemini 3.1-Pro
DeepSeek V4-Pro-Max在标准推理基准测试中超越了GPT-5.2和Gemini 3.0-Pro，这表明了开源模型在性能上的巨大提升。
6 Matching Annotations
- May 2026
-
nlp.elvissaravia.com nlp.elvissaravia.com
-
- Apr 2026
-
www.technologyreview.com www.technologyreview.com
-
DeepSeek V4 exceeds them all on coding, math, and STEM problems, making it one of the strongest open-source models ever released.
大多数人认为开源AI模型在性能上无法匹敌闭源商业模型，但作者认为DeepSeek V4在多个关键领域超越了其他开源模型，甚至与顶级闭源模型相当。这挑战了'开源必然意味着性能妥协'的行业共识，暗示开源模型正在迅速缩小与商业模型的差距。
-
-
www.tomtunguz.com www.tomtunguz.com
-
Gemma 4 E4B matches or exceeds GPT-4o across multiple benchmarks including MATH, GSM8K, GPQA Diamond & HumanEval
这一性能对比结果令人惊讶，表明开源模型已经能够闭源模型的性能，这可能打破AI领域的封闭生态，促进更广泛的研究合作和创新，同时降低企业采用AI的门槛。
Tags
Annotators
URL
-
-
-
Byte for byte, the most capable open models
大多数人认为开源模型在性能上无法与闭源/专有模型相提并论，但作者声称Gemma 4是'字节对字节最强大的开源模型'，挑战了这一行业共识。这暗示开源模型在特定指标上已经超越了商业闭源模型，是一个非传统的观点。
-
- Apr 2025
-
esbuild.github.io esbuild.github.io
-
Above: the time to do a production bundle
Nice way to demonstrate and let people feel how slow the competition is!
-
- Mar 2021
-
-
McCabe, Stefan, Leo Torres, Timothy LaRock, Syed Arefinul Haque, Chia-Hung Yang, Harrison Hartle, and Brennan Klein. ‘Netrd: A Library for Network Reconstruction and Graph Distances’. ArXiv:2010.16019 [Physics], 29 October 2020. http://arxiv.org/abs/2010.16019.
-