#llm search results

Agentic AI ≠ chatbot. Agentic AI ≠ RAG. Agentic AI ≠ RPA + LLM. Real Agentic AI includes: • Goal ownership • Planning • Memory • Tool use • Feedback loops • Multi-agent coordination It executes objectives, not just prompts. 🧠 #AgenticAI #LLM #AIArchitecture

ingliguori's tweet image. Agentic AI ≠ chatbot.
Agentic AI ≠ RAG.
Agentic AI ≠ RPA + LLM.
Real Agentic AI includes:
• Goal ownership
• Planning
• Memory
• Tool use
• Feedback loops
• Multi-agent coordination
It executes objectives, not just prompts. 🧠
#AgenticAI #LLM #AIArchitecture

韓国ネイバーの強みは、ローカル文脈への深い理解とコミュニティ性。グローバルLLMも、特定の地域・文化に特化した学習データでチューニングすれば、驚くほど精度を高められる。検索の質は、基盤モデルの汎用性だけじゃない。#LLM


7 Layered of LLM Stack 📘📚 #ai #llm

dkare1009's tweet image. 7 Layered of LLM Stack 📘📚 

#ai #llm

推理速度提升最高 6 倍。DFlash 技术通过 Block Diffusion 模型实现多词并行预测,在 Qwen3.5 等最新模型上可达 400+ tokens/s。相比现有顶尖方法 EAGLE-3,其速度提升最高达 2.5 倍,且能保持模型精度,可降低服务器成本并优化用户体验。#LLM #DFlash


I built an OS that solves these structural pains. Try the live demo right now (GitHub Pages): naoyukioyama561-alt.github.io/SHI-Claude-Con… #Claude #AI #LLM #AIGovernance #FailureModes


“Ever wondered what makes an #LLM really tick—what’s behind the colorful pinball-machine metaphors?” Learn more about #parameters here: buff.ly/5tzH5tu #TechNews

TheOfficialACM's tweet image. “Ever wondered what makes an #LLM really tick—what’s behind the colorful pinball-machine metaphors?”

Learn more about #parameters here: buff.ly/5tzH5tu #TechNews

002: Agentic space + LLMs? Mathias shares fascinating project insights. The future is bright! #AI #LLM


AgentLLM.co • Short & memorable • Perfect for AI agents niche • Strong keyword (LLM) Ideal for startups building in AI #AI #ArtificialIntelligence #LLM #AIAgents #MachineLearning #SaaS

domainTopseller's tweet image. AgentLLM.co

• Short & memorable
• Perfect for AI agents niche
• Strong keyword (LLM)

Ideal for startups building in AI
#AI #ArtificialIntelligence #LLM
#AIAgents
#MachineLearning
#SaaS

Claude is iterating at a remarkable pace. Given the frequency of releases, I’ve aggregated most of the new features and capabilities announced thus far. #Anthropic #AI #LLM #GenerativeAI #ProductUpdates

fateedex_mac's tweet image. Claude is iterating at a remarkable pace. Given the frequency of releases, I’ve aggregated most of the new features and capabilities announced thus far.

 #Anthropic #AI #LLM #GenerativeAI #ProductUpdates
fateedex_mac's tweet image. Claude is iterating at a remarkable pace. Given the frequency of releases, I’ve aggregated most of the new features and capabilities announced thus far.

 #Anthropic #AI #LLM #GenerativeAI #ProductUpdates
fateedex_mac's tweet image. Claude is iterating at a remarkable pace. Given the frequency of releases, I’ve aggregated most of the new features and capabilities announced thus far.

 #Anthropic #AI #LLM #GenerativeAI #ProductUpdates
fateedex_mac's tweet image. Claude is iterating at a remarkable pace. Given the frequency of releases, I’ve aggregated most of the new features and capabilities announced thus far.

 #Anthropic #AI #LLM #GenerativeAI #ProductUpdates

You don’t need a better LLM. You’re just using it badly. InferScale 0.1.2: → Generate multiple outputs → Score them → Pick the best Now faster with batch inference. Same model. Better results. Most people aren’t ready for this shift. github.com/mbaddar1/Infer… #AI #LLM


We compressed LLM KV cache by 8× at 1-bit with ×12 better distortion than SOTA (TurboQuant). No fine-tuning. No architecture change. Drop-in on any Transformer. Fully reproducible benchmark. Method: patent pending huggingface.co/spaces/Nandu97… #LLM #efficiency #research


AIニュースダイジェスト ・主要LLMの性能競争が激化、Anthropicの「Opus 4.7」が躍進 ・AIコンパニオンの会話履歴を基にローカルモデルをファインチューニングするツール登場 ・AnthropicとOpenAI、AI設計アプローチで異なる哲学 #AIニュース #AI #LLM


The latest Karlsen AI-AMD version is now available on GitHub, please check it out. github.com/karlsencoin/ka#GPU #AI #LLM #Private #Anonymous


I have identified the core of AI safety risks: Paradox Closure Mechanism. Conflicts between objective facts and subjective beliefs may lead to unstable logic loops. Detecting such loops can make AI more reliable. This could be a foundation for AI safety. #AISafety #AGI #LLM


Ever had an AI forget the plan halfway? That’s the Context Window limit. LLMs have a "working memory." Once it’s full, old data gets trashed for the new. Like a 10-page desk—to read page 11, page 1 goes in the bin. Mastering this = AI Engineering. 🛠️ #GenAI #Engineering #LLM


#llm #AI Já cheguei a ter 95k downloads em um mês. Quando vejo algo interessante posto algo novo. huggingface.co/Brunobkr


Excited to share my research on a Hybrid LLM Routing System.• Cut API usage by 46.7% • Retained 94% API accuracy on GSM8KKey insight: LLM self-confidence is poorly calibrated → better routing is needed. #AI #LLM #BuildInPublic

SARTHAK27675014's tweet image. Excited to share my research on a Hybrid LLM Routing System.• Cut API usage by 46.7%
• Retained 94% API accuracy on GSM8KKey insight: LLM self-confidence is poorly calibrated → better routing is needed.
#AI #LLM #BuildInPublic
SARTHAK27675014's tweet image. Excited to share my research on a Hybrid LLM Routing System.• Cut API usage by 46.7%
• Retained 94% API accuracy on GSM8KKey insight: LLM self-confidence is poorly calibrated → better routing is needed.
#AI #LLM #BuildInPublic

Seedance 2.0 just turned video AI into a meme weapon. From viral Star Wars remixes to eerily real apology videos, X is losing it. Is this another hype cycle — or the moment video LLMs went mainstream? 👇 #Seedance2 #Meme #LLM #Videos #Shorts #AI spaisee.com/seedance-2-0-a…


Flying under the spotlight is EvoCUA. This open-source multimodal model tackles end-to-end computer automation tasks, achieving top-tier results on the OSWorld benchmark with strong cross-OS generalization capabilities on Windows. #OSS #LLM hublens.dev/en/meituan-Evo…


Is your AI being poisoned? Madhuri Chandoor from PromptHalo reveals how malicious code on websites can drift and compromise LLMs. Don't miss this critical look at AI security risks. #AI #LLM #CyberSecurity #PromptHalo


🔄 AI model failover isn't just a nice-to-have—it's essential for production systems. When one model is down or slow, intelligent routing keeps your applications running smoothly. #AIReliability #LLM #TechInfrastructure #AIStrategy #SystemDesign


#LLM VS #RAG RAG VS #Al Agent VS Agentic Al

Eease11's tweet image. #LLM VS #RAG
RAG VS #Al Agent VS Agentic Al

“Which LLM is best?” 🤔 Wrong question. Bench AI → run one prompt across models + compare outputs side-by-side. See behavior, not just “winners.” Code → github.com/darkrishabh/be… Live (hosted on @vercel) → bench-ai-web.vercel.app #AI #LLM #Claude #Codex #GPT #OpenSource


✅️ #ai #llm #gpt

nwbotz's tweet image. ✅️ #ai #llm #gpt

I'm not surprised. The wealthy are taking over this space, just like they do with financial markets (e.g., Bloomberg terminals). GPT-4 will be considered ghetto AI in the next 3-4 years, if not sooner. GPT-6, 7, 8, and 9 will be completely controlled by the 1% and used to gain



I am amazed with claude design skills. Just used 1st tym and it literally blow my mind, gives fucking amazing design Great work by @claudeai team #claude #llm #claudedesign


LLM Quantization @OpenAI #destructed it's #LLM model by #quantizing it into an #SLM (#Small #Language #Model) while they #restricted their model for unbias liberties By doing so they destroyed the #DALLE3 fully and even #reasoning is affected to the point of #ridculousness

My boss told me to make some slides on GPT-5. I didn’t want to... but here it is. chat.z.ai/space/t0up16h9… #GPT5

Zai_org's tweet image. My boss told me to make some slides on GPT-5. I didn’t want to... but here it is.
chat.z.ai/space/t0up16h9…

#GPT5


The Future of #LLM #Chat is Multilingual. Own it Today. #Chatgpt Break the language barrier with Chatafied.com. We’ve built a high-performance #AIchat app that scales globally. Now, we’re looking for the right owner to take it to the masses. Seamless #AI

altcoingazette's tweet image. The Future of #LLM #Chat is Multilingual. Own it Today.  #Chatgpt

Break the language barrier with Chatafied.com. 

We’ve built a high-performance #AIchat app that scales globally. Now, we’re looking for the right owner to take it to the masses.  Seamless #AI

"I build turnkey #AI apps so you don't have to. 🛠️ From #LLM, #AI, #Agentic Data tools to Multilingual #Chat, the #Zysparq portfolio is now open for acquisition. Production-ready. Clean #code. Own your next #SaaS today: zysparqtechnologies.com/ai-apps-for-sa…

altcoingazette's tweet image. "I build turnkey #AI apps so you don't have to. 🛠️

From #LLM, #AI, #Agentic Data tools to Multilingual #Chat, the #Zysparq portfolio is now open for acquisition. Production-ready. Clean #code. 

Own your next #SaaS today: 
zysparqtechnologies.com/ai-apps-for-sa…

Loaded our entire codebase into one context window. 10 million tokens. Zero API bill. Llama 4 is quietly changing who can afford to build. #AgenticAI #LLM


Agentic AI ≠ chatbot. Agentic AI ≠ RAG. Agentic AI ≠ RPA + LLM. Real Agentic AI includes: • Goal ownership • Planning • Memory • Tool use • Feedback loops • Multi-agent coordination It executes objectives, not just prompts. 🧠 #AgenticAI #LLM #AIArchitecture

ingliguori's tweet image. Agentic AI ≠ chatbot.
Agentic AI ≠ RAG.
Agentic AI ≠ RPA + LLM.
Real Agentic AI includes:
• Goal ownership
• Planning
• Memory
• Tool use
• Feedback loops
• Multi-agent coordination
It executes objectives, not just prompts. 🧠
#AgenticAI #LLM #AIArchitecture

“Ever wondered what makes an #LLM really tick—what’s behind the colorful pinball-machine metaphors?” Learn more about #parameters here: buff.ly/5tzH5tu #TechNews

TheOfficialACM's tweet image. “Ever wondered what makes an #LLM really tick—what’s behind the colorful pinball-machine metaphors?”

Learn more about #parameters here: buff.ly/5tzH5tu #TechNews

Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation Bae et al.: arxiv.org/abs/2507.10524 #ArtificialIntelligence #AIAgents #LLM

ceobillionaire's tweet image. Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation

Bae et al.: arxiv.org/abs/2507.10524

#ArtificialIntelligence #AIAgents #LLM

"query fan-outs", lesquelles peut-on observer ? Via @Moz #LLM #DigitalMarketing

AnthonyRochand's tweet image. "query fan-outs", lesquelles peut-on observer ?

Via @Moz

#LLM #DigitalMarketing

Claude Code opened to any LLM — OpenAI, Gemini, DeepSeek, Ollama, and 200+ models via OpenAI-compatible API shim #Claude #Code #LLM #OpenAI #Gemini #DeepSeek #Ollama

Python_Dv's tweet image. Claude Code opened to any LLM — OpenAI, Gemini, DeepSeek, Ollama, and 200+ models via OpenAI-compatible API shim

#Claude #Code #LLM #OpenAI #Gemini #DeepSeek #Ollama

PiTorch: machine learning on bare metal Raspberry Pi Zeros #LLM @Raspberry_Pi blog.adafruit.com/2026/04/14/pit…

adafruit's tweet image. PiTorch: machine learning on bare metal Raspberry Pi Zeros  #LLM @Raspberry_Pi 

blog.adafruit.com/2026/04/14/pit…

Loading...

Something went wrong.


Something went wrong.