Apr 26, 2026, 10:40 p.m.·小冬瓜AIGCscore 77.8modelsllm-systemsmodels
deep dive into DeepSeek-V4 (1) : model architecture
小冬瓜AIGC published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle2 upvotes
View Chinese source card → · Original source
Apr 25, 2026, 11:10 a.m.·果冻虾仁score 69.6inferencemodelsllm-systemsoperator
Chinese technical note: 如何Day 0fast deploymentopen-source LLMs?
果冻虾仁 published a Zhihu article relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Apr 14, 2026, 7:22 a.m.·吃果冻不吐果冻皮score 80.4agentsinferencemodelsresearch
Chinese technical note: SkillReducer:为Skills瘦身,破解Token低效难题
吃果冻不吐果冻皮 published a Zhihu article relevant to AI agents and coding workflows, inference and AI systems, frontier and open model development, research signals. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Apr 12, 2026, 4:23 a.m.·skydownacaiscore 68.3post-trainingagentsmodelsevals
Chinese technical note: Claude Code 模型 RLtraining中的Reward Hacking
skydownacai published a Zhihu article relevant to post-training and RL, AI agents and coding workflows, frontier and open model development, evaluation and reliability. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle32 upvotes
View Chinese source card → · Original source
Apr 7, 2026, 9:31 p.m.·金雪锋score 71.8agentsagentsmodelsresearch
Chinese technical note: 北向agents化,将重新定义AI Infra的易用性和生态
金雪锋 published a Zhihu article relevant to AI agents and coding workflows. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Apr 3, 2026, 9:50 p.m.·金雪锋score 71.8researchagentsmodelsresearch
Chinese technical note: 超节点亲和AI框架
金雪锋 published a Zhihu article relevant to research signals. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle1 upvotes
View Chinese source card → · Original source
Apr 3, 2026, 10:02 a.m.·紫气东来score 89.6agentsmodelsllm-systemsmodels
Chinese technical note: 【招人帖-实习】腾讯微信基础-LLMs算法研究员
紫气东来 published a Zhihu article relevant to AI agents and coding workflows, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Feb 22, 2026, 9:32 a.m.·字节score 74.1agentsagentsmodels
Chinese technical note: Vibe Engineering 进化史:三重门背后的思维跃迁
字节 published a Zhihu article relevant to AI agents and coding workflows. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle3 upvotes
View Chinese source card → · Original source
Feb 17, 2026, 3:50 p.m.·兽族机枪兵score 71.7agentsmodelsresearchagents
Chinese technical note: Kimi K2.5 technical notes有点多啊
兽族机枪兵 published a Zhihu article relevant to AI agents and coding workflows, frontier and open model development, research signals. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle1 upvotes
View Chinese source card → · Original source
Feb 16, 2026, 12:40 p.m.·果冻虾仁score 69.6inferencemodelsllm-systemsoperator
Chinese technical note: 管中窥豹nano-vllm(三):ModelRunner补充
果冻虾仁 published a Zhihu article relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Feb 16, 2026, 2:13 a.m.·兽族机枪兵score 71.7inferenceagentsmodels
Chinese technical note: 投机采样方案里,EAGLE-3是如何做到一块显卡同时deployment目标模型和草稿模型的?
兽族机枪兵 published a Zhihu article relevant to inference and AI systems. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Feb 15, 2026, 6:35 a.m.·兽族机枪兵score 71.7modelsagentsmodels
Chinese technical note: 终于有时间好好看看DeepSeek-OCR了。。
兽族机枪兵 published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Feb 7, 2026, 12:14 p.m.·果冻虾仁score 69.6inferencemodelsllm-systemsoperator
Chinese technical note: vllm使用plugin集成外部模型结构
果冻虾仁 published a Zhihu article relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle1 upvotes
View Chinese source card → · Original source
Jan 17, 2026, 8:23 a.m.·紫气东来score 89.6modelsllm-systemsmodels
Chinese technical note: LLM(34):Batch size 与 Learning rate 的理论与实践
紫气东来 published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle2 upvotes
View Chinese source card → · Original source
Jan 13, 2026, 9:27 p.m.·小冬瓜AIGCscore 77.8modelsllm-systemsmodels
Chinese technical note: 【deep dive intoEngram】DeepSeek 的 Conditional Memory 能取代 Attention 吗?(超长文、附代码)
小冬瓜AIGC published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle12 upvotes
View Chinese source card → · Original source
Jan 7, 2026, 2:48 a.m.·金雪锋score 71.8agentsinferenceresearchagents
Chinese technical note: AKG kernel agents:利用multi-agents进行kernel的生成和迁移
金雪锋 published a Zhihu article relevant to AI agents and coding workflows, inference and AI systems, research signals. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Jan 4, 2026, 5:57 a.m.·字节score 74.1agentsagentsmodels
Chinese technical note: 深读 Manus:寻找 agents 产品的“品味”与“定力”
字节 published a Zhihu article relevant to AI agents and coding workflows. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle10 upvotes
View Chinese source card → · Original source
Jan 2, 2026, 6:29 p.m.·小冬瓜AIGCscore 77.8modelsllm-systemsmodels
Chinese technical note: 【deep dive into mHC】详解DeepSeek残差链接mHC进化之路(超长文、附代码)
小冬瓜AIGC published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle31 upvotes
View Chinese source card → · Original source
Dec 31, 2025, 1:36 a.m.·小冬瓜AIGCscore 77.8post-trainingmodelsllm-systemsmodels
Chinese technical note: 做正确且难的事——小冬瓜AIGC的25年终总结
小冬瓜AIGC published a Zhihu article relevant to post-training and RL, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Dec 30, 2025, 9:26 a.m.·吃果冻不吐果冻皮score 80.4modelsevalsllm-systemsmodels
Chinese technical note: 你真的搞懂了LLM性能压测的各项指标吗?
吃果冻不吐果冻皮 published a Zhihu article relevant to frontier and open model development, evaluation and reliability. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Dec 4, 2025, 2:15 a.m.·skydownacaiscore 68.3post-trainingagentsmodelspost-training
Chinese technical note: LLM agents RL的一些实践感悟
skydownacai published a Zhihu article relevant to post-training and RL, AI agents and coding workflows, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle23 upvotes
View Chinese source card → · Original source
Nov 25, 2025, 6:21 a.m.·吃果冻不吐果冻皮score 80.4modelsllm-systemsmodels
Chinese technical note: DeepSeek 视觉语言LLMs技术演进(从DeepSeek VL/VL2到DeepSeek OCR)
吃果冻不吐果冻皮 published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Nov 14, 2025, 3:04 a.m.·xhchenscore 97.2modelspost-trainingmodelsresearch
Chinese technical note: multimodalLLMs入门:举一个例子说明SW-MSA
xhchen published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Oct 25, 2025, 5:57 a.m.·吃果冻不吐果冻皮score 80.4inferencemodelsllm-systemsmodels
Chinese technical note: 谈谈LLM生成文本的惩罚参数
吃果冻不吐果冻皮 published a Zhihu article relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle1 upvotes
View Chinese source card → · Original source
Oct 25, 2025, 3:42 a.m.·vibe memoscore 82.7modelsagentsmodels
Chinese technical note: DeepSeek团队发布视觉压缩OCR模型,哪些信息和技术亮点值得关注?
vibe memo published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Oct 25, 2025, 3:42 a.m.·vibe memoscore 82.7modelsagentsmodels
Chinese technical note: DeepSeek-OCR与Glyph: 视觉压缩会重塑LLM的输入范式?
vibe memo published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Oct 13, 2025, 10:24 a.m.·吃果冻不吐果冻皮score 80.4inferencemodelsllm-systemsmodels
Chinese technical note: How to read Thinking Machines Lab发表的首篇长文《克服 LLM inference中的不确定性》?
吃果冻不吐果冻皮 published a Zhihu answer relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Sep 30, 2025, 8:48 a.m.·skydownacaiscore 68.3post-trainingagentsmodels
Chinese technical note: RL崩溃背后的训推mismatch
skydownacai published a Zhihu article relevant to AI engineering. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle4 upvotes
View Chinese source card → · Original source
Sep 19, 2025, 10:18 a.m.·vibe memoscore 82.7agentsevalsagentsmodels
Tongyi DeepResearch series: WebWeaver, WebResearcher, ReSum, WebSailor-V2,agentsScaler
vibe memo published a Zhihu article relevant to AI agents and coding workflows, evaluation and reliability. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Sep 19, 2025, 3:08 a.m.·字节score 74.1agentsmodelsagentsmodels
Chinese technical note: 让agents“记得对、做得准、跑得快”:Context Engineering实战
字节 published a Zhihu article relevant to AI agents and coding workflows, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle13 upvotes
View Chinese source card → · Original source
Sep 16, 2025, 8:14 a.m.·vibe memoscore 82.7agentsmodelsagentsmodels
Chinese technical note: Writing effective tools for agentss — with agentss
vibe memo published a Zhihu article relevant to AI agents and coding workflows, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Sep 14, 2025, 5:09 a.m.·skydownacaiscore 68.3post-trainingpost-trainingagentsmodels
Chinese technical note: RLtraining中Why熵减往往意味着training收敛?
skydownacai published a Zhihu article relevant to post-training and RL. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle4 upvotes
View Chinese source card → · Original source
Sep 10, 2025, 6:33 a.m.·紫气东来score 89.6post-trainingagentsmodelsllm-systems
Chinese technical note: 【招人帖】腾讯微信基础LLMs算法
紫气东来 published a Zhihu article relevant to post-training and RL, AI agents and coding workflows, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle1 upvotes
View Chinese source card → · Original source
Sep 9, 2025, 9:52 p.m.·vibe memoscore 82.7inferencemodelsagentsmodels
Chinese technical note: 深入 vLLM:高吞吐量 LLM inference系统的剖析
vibe memo published a Zhihu article relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Aug 22, 2025, 4:08 a.m.·xhchenscore 97.2post-trainingpost-trainingmodelsresearch
Chinese technical note: 在SFTtraining阶段,有大量短文本数据与少量的长文本数据,从training效率和模型性能的角度,应该如何选择批处理策略?
xhchen published a Zhihu article relevant to post-training and RL. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Aug 19, 2025, 1:19 a.m.·xhchenscore 97.2modelspost-trainingmodelsresearch
Chinese technical note: 主流LLMs中的“混合结构”
xhchen published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Aug 13, 2025, 5:59 a.m.·xhchenscore 97.2modelsresearchpost-trainingmodels
Linear Attention vs Self Attention
xhchen published a Zhihu article relevant to frontier and open model development, research signals. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle0 upvotes
View Chinese source card → · Original source
Aug 12, 2025, 12:31 a.m.·字节score 74.1agentsinferencemodelsagents
Chinese technical note: 烧掉上亿 Token 后,我总结的 Coding agents 高阶玩法
字节 published a Zhihu article relevant to AI agents and coding workflows, inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle12 upvotes
View Chinese source card → · Original source
Aug 5, 2025, 10:22 p.m.·小冬瓜AIGCscore 77.8modelsllm-systemsmodels
Chinese technical note: How to read OpenAI 8 月 6 日推出的开源模型 gpt-oss?和国产开源模型相比怎么样?
小冬瓜AIGC published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer2 upvotes
View Chinese source card → · Original source
Jul 23, 2025, 9:16 a.m.·紫气东来score 89.6post-trainingmodelsllm-systemsmodels
Chinese technical note: Reasoning LLM(六):内在reward
紫气东来 published a Zhihu article relevant to post-training and RL, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle2 upvotes
View Chinese source card → · Original source
Jun 30, 2025, 1:39 a.m.·南爸安爸score 72.2inferencemodelsresearchllm-systems
Chinese technical note: Meta 推荐芯片MTIAV2分析(ISCA 25)
南爸安爸 published a Zhihu article relevant to inference and AI systems, frontier and open model development, research signals. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle1 upvotes
View Chinese source card → · Original source
Jun 27, 2025, 12:55 a.m.·吃果冻不吐果冻皮score 80.4modelsllm-systemsmodels
Chinese technical note: LLMs中目前最快最好的解码策略What is呢?还有哪些问题值得研究?
吃果冻不吐果冻皮 published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer1 upvotes
View Chinese source card → · Original source
Jun 14, 2025, 12:11 p.m.·字节score 74.1agentsagentsmodels
Chinese technical note: OpenAI 上线新一代coding神器 Codex,有哪些技术亮点?程序员的工作会被彻底颠覆吗?
字节 published a Zhihu answer relevant to AI agents and coding workflows. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Jun 5, 2025, 12:25 a.m.·小冬瓜AIGCscore 77.8llm-systemsmodels
Chinese technical note: MoEtraining中的Top-K运算不会导致不可导(不连续)吗?
小冬瓜AIGC published a Zhihu answer relevant to AI engineering. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer16 upvotes
View Chinese source card → · Original source
May 29, 2025, 11:32 a.m.·方佳瑞score 74.2post-trainingllm-systemsmodelsoperator
Chinese technical note: 重读 Google 旧文 Pathways,寻找 veRL 中 Single-controller 思想源头
方佳瑞 published a Zhihu article relevant to post-training and RL. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle1 upvotes
View Chinese source card → · Original source
May 24, 2025, 12:02 p.m.·字节score 74.1modelsagentsmodels
Chinese technical note: ChatGPT最实用的提示(Prompts)写法有哪些?
字节 published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer3 upvotes
View Chinese source card → · Original source
Apr 27, 2025, 10:56 a.m.·方佳瑞score 74.2post-trainingmodelsllm-systemsmodels
Chinese technical note: 字节豆包LLMs团队与香港大学发布全新 RLHF 框架提升 1.5~20 倍吞吐量,这意味着什么?
方佳瑞 published a Zhihu answer relevant to post-training and RL, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer4 upvotes
View Chinese source card → · Original source
Apr 27, 2025, 10:56 a.m.·方佳瑞score 74.2post-trainingmodelsllm-systemsmodels
Chinese technical note: veRL:All in RL元年的必修课
方佳瑞 published a Zhihu article relevant to post-training and RL, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle20 upvotes
View Chinese source card → · Original source
Apr 16, 2025, 9:19 a.m.·方佳瑞score 74.2inferencemodelsllm-systemsmodels
Chinese technical note: SageAttention:即插即用的8-bit Attention 最佳实践
方佳瑞 published a Zhihu article relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle9 upvotes
View Chinese source card → · Original source
Mar 23, 2025, 3:12 a.m.·南爸安爸score 72.2inferencemodelsllm-systemsmodels
Chinese technical note: How to readNvidia发布的LLMsinferencePD分离architectureDynamo?
南爸安爸 published a Zhihu answer relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Feb 20, 2025, 3:49 a.m.·吃果冻不吐果冻皮score 80.4inferencemodelsllm-systemsmodels
Chinese technical note: 百度旗下 AI 芯片昆仑芯支持单机deployment DeepSeek 满血版LLMs,此举有何深意?
吃果冻不吐果冻皮 published a Zhihu answer relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer1 upvotes
View Chinese source card → · Original source
Feb 16, 2025, 10:56 a.m.·方佳瑞score 74.2inferencemodelsllm-systemsmodels
Chinese technical note: DeepSeek-R1 API定价Why这么便宜,在目前全量模型deployment非常困难的情况下,是否注定亏损?
方佳瑞 published a Zhihu answer relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer5 upvotes
View Chinese source card → · Original source
Feb 16, 2025, 10:56 a.m.·方佳瑞score 74.2inferencemodelsllm-systemsmodels
Chinese technical note: 吃瓜DeepSeek inference成本需要的相关概念:Throughput,TPOT,TTFT
方佳瑞 published a Zhihu article relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle7 upvotes
View Chinese source card → · Original source
Feb 1, 2025, 3:22 a.m.·南爸安爸score 72.2modelsllm-systemsmodels
Chinese technical note: 为何腾讯、阿里、华为等大厂难出DeepSeek级颠覆产品?是组织僵化还是创新无能?
南爸安爸 published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Jan 21, 2025, 1:45 a.m.·紫气东来score 89.6modelsllm-systemsmodels
Chinese technical note: How to read DeepSeek 的 R1 与 R1-Zero 模型?
紫气东来 published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer2 upvotes
View Chinese source card → · Original source
Jan 21, 2025, 12:48 a.m.·小冬瓜AIGCscore 77.8modelsllm-systemsmodels
Chinese technical note: How to readdeepseek预发布的deepseek-R1?
小冬瓜AIGC published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Jan 20, 2025, 10:02 p.m.·skydownacaiscore 68.3modelspost-trainingagentsmodels
Chinese technical note: How to read DeepSeek 的 R1 与 R1-Zero 模型?
skydownacai published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer8 upvotes
View Chinese source card → · Original source
Jan 16, 2025, 3:15 a.m.·紫气东来score 89.6llm-systemsmodels
Chinese technical note: How to read MiniMax 于 2025 年 1 月 15 日发布的 MiniMax-01 系列模型?
紫气东来 published a Zhihu answer relevant to AI engineering. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer2 upvotes
View Chinese source card → · Original source
Jul 26, 2024, 9:27 p.m.·紫气东来score 89.6inferencellm-systemsmodels
Chinese technical note: 可否给一个实际例子,结合GPU硬件architecture和Transformerarchitecture讲一讲,如何提高training和inference速度?
紫气东来 published a Zhihu answer relevant to inference and AI systems. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Jul 20, 2024, 8:57 a.m.·vibe memoscore 82.7modelsagentsmodels
Chinese technical note: 阿里云Qwen2两小时登顶HuggingFaceopen-source LLMs榜首,你怎么看?
vibe memo published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Jul 10, 2024, 10:12 p.m.·九号score 70.3inferencemodels
Chinese technical note: 单卡可Million-contextinferenceTTFT 10倍加速 - MInference 1.0
九号 published a Zhihu article relevant to inference and AI systems. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle20 upvotes
View Chinese source card → · Original source
Jul 10, 2024, 8:01 a.m.·小冬瓜AIGCscore 77.8modelsllm-systemsmodels
Chinese technical note: Why加速LLM推断有KV Cache而没有Q Cache?
小冬瓜AIGC published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer1 upvotes
View Chinese source card → · Original source
Jun 26, 2024, 10:15 a.m.·vibe memoscore 82.7inferencemodelsagentsmodels
Chinese technical note: LLMsinference优化哪些公司或者团队做的好?
vibe memo published a Zhihu answer relevant to inference and AI systems, frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Jun 23, 2024, 11:45 p.m.·vibe memoscore 82.7modelsagentsmodels
Chinese technical note: trainingmultimodal模型有哪些trick?
vibe memo published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer1 upvotes
View Chinese source card → · Original source
Jun 17, 2024, 10:52 p.m.·九号score 70.3modelsevalsmodels
Chinese technical note: 请问现在有哪些研究和数据集可以evaluation大语言模型llm的长文本理解能力?
九号 published a Zhihu answer relevant to frontier and open model development, evaluation and reliability. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer0 upvotes
View Chinese source card → · Original source
Jun 17, 2024, 10:52 p.m.·九号score 70.3modelsevalsmodels
Chinese technical note: 全面evaluationLLMs长文本能力的Benchmarks
九号 published a Zhihu article relevant to frontier and open model development, evaluation and reliability. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle1 upvotes
View Chinese source card → · Original source
Mar 23, 2024, 3:34 p.m.·ZZZZJscore 74.3evalsresearchllm-systems
Chinese technical note: 读 Mora 论文笔记
ZZZZJ published a Zhihu article relevant to evaluation and reliability, research signals. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle4 upvotes
View Chinese source card → · Original source
Mar 4, 2024, 7:25 p.m.·九号score 70.3modelsmodels
Chinese technical note: 如何看待Anthropic公司在ChatGPT4.5推出前,宣布推出Claude 3?
九号 published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer4 upvotes
View Chinese source card → · Original source
Feb 25, 2024, 11:59 p.m.·九号score 70.3models
Chinese technical note: 谷歌正式推出开源大语言模型 Gemma,声称超越 Meta Llama-2 竞品,将带来哪些影响?
九号 published a Zhihu answer relevant to AI engineering. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer3 upvotes
View Chinese source card → · Original source
Feb 25, 2024, 11:59 p.m.·九号score 70.3modelsevalsmodels
Chinese technical note: evaluationGemma - 对比Mistral,Qwen1.5
九号 published a Zhihu article relevant to frontier and open model development, evaluation and reliability. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle11 upvotes
View Chinese source card → · Original source
Feb 19, 2024, 7:50 p.m.·九号score 70.3models
Chinese technical note: 可以用大语言模型评估大语言模型的结果吗?
九号 published a Zhihu answer relevant to AI engineering. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer5 upvotes
View Chinese source card → · Original source
Feb 19, 2024, 7:50 p.m.·九号score 70.3modelsevalsresearchmodels
Chinese technical note: 出「月考卷」来杜绝LLMs在evaluation中的作弊
九号 published a Zhihu article relevant to frontier and open model development, evaluation and reliability, research signals. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle8 upvotes
View Chinese source card → · Original source
Sep 24, 2023, 12:03 a.m.·南爸安爸score 72.2modelsllm-systemsmodels
Chinese technical note: 万亿LLM/multimodal MoE和万亿EmbeddingLLMs系统设计上的共性和差异的一点体会
南爸安爸 published a Zhihu article relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuarticle10 upvotes
View Chinese source card → · Original source
Aug 12, 2023, 8:47 a.m.·兽族机枪兵score 71.7agentsmodels
Chinese technical note: WhyNLP模型通常使用AdamW作为优化器,而不是SGD?
兽族机枪兵 published a Zhihu answer relevant to AI engineering. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer27 upvotes
View Chinese source card → · Original source
Aug 4, 2023, 5:00 a.m.·兽族机枪兵score 71.7modelsagentsmodels
Chinese technical note: 你会花3w学chatgpt和mid journey,然后就业吗?
兽族机枪兵 published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer3 upvotes
View Chinese source card → · Original source
Mar 27, 2023, 12:34 p.m.·ZZZZJscore 74.3inferencellm-systems
Chinese technical note: 模型inferenceCUDA100% GPU0%What is情况?
ZZZZJ published a Zhihu answer relevant to inference and AI systems. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer6 upvotes
View Chinese source card → · Original source
Mar 19, 2023, 3:55 a.m.·金雪锋score 71.8modelsagentsmodelsresearch
Chinese technical note: ChatGPT的出现对哲学会产生哪些挑战?
金雪锋 published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer46 upvotes
View Chinese source card → · Original source
Dec 19, 2022, 11:38 p.m.·skydownacaiscore 68.3post-trainingpost-trainingagentsmodels
Chinese technical note: 求:复旦reinforcement learning老师推荐?
skydownacai published a Zhihu answer relevant to post-training and RL. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer4 upvotes
View Chinese source card → · Original source
Dec 11, 2022, 8:51 p.m.·金雪锋score 71.8modelsagentsmodelsresearch
Chinese technical note: How to read OpenAI 的超级对话模型 ChatGPT ?
金雪锋 published a Zhihu answer relevant to frontier and open model development. The original Chinese excerpt is included below so the feed can preserve the raw source while giving English readers enough context to triage the item.
Zhihuanswer4 upvotes
View Chinese source card → · Original source