Learn everything there is to know about LlamaParse in this comprehensive video! In this video, @mesudarshan covers: ➡️ Multiple parsing modes ➡️ Using parsing instructions to improve quality ➡️ Output formats available ➡️ Parsing audio and images ➡️ JSON mode ➡️ Using it all in https://x.com/llama_index/status/1890499579214491967
stepfun-ai/
“Github 👨🔧: Framework for building scalable agentic applications. → Open-source TypeScript library for building production-ready multi-agent systems. → Provides production-grade features including memory management, agent state serialization, and secure sandboxed code https://x.com/rohanpaul_ai/status/1890550582999445686
“🚀 LangChain in Atlanta! 🚀 Join us next Thursday, February 27th for an evening of AI at the Honeywell offices with LangChain CEO, @hwchase17” / X https://x.com/LangChainAI/status/1892301688138047842
“We’ve just released the coolest feature ever in smolagents: you can now share agents to the Hub! 🥳🥳 And any agent pushed to Hub get a cool Space interface to directly chat with it. This was a real technical challenge: for instance, serializing tools to export them meant that https://x.com/AymericRoucher/status/1890431468700332366
“DeepSeek R1 was just the start—this new Chinese research from @Kimi_Moonshot lets RAG AI agents devour entire codebases and documentation with no context limits. Mixture of Experts and Sparse attention make near-infinite context possible. 🧵1/n 📌 Challenge of Long-Context https://x.com/rohanpaul_ai/status/1892535262879617101
“OmniParser, groundbreaking screenshot parser for web automation, got even better and faster 🔥 It’s open-source (MIT), and you can plug this in model/agent of your choice: Qwen2.5VL, DeepSeek R1, 4o/o1/o3 mini Model and demo out on @huggingface https://x.com/mervenoyann/status/1891524621435830700
“AlphaMaze: Teaching a 1.5B LLM to think visually and solve ARC-AGI like puzzles! 🤯 Powered by DeepSeek R1 1.5B + GRPO All with Apache licensed checkpoints and dataset 🤗 https://x.com/reach_vb/status/1892999150255440012
“🚀 Day 0: Warming up for #OpenSourceWeek! We’re a tiny team @deepseek_ai exploring AGI. Starting next week, we’ll be open-sourcing 5 repos, sharing our small but sincere progress with full transparency. These humble building blocks in our online service have been documented,” / X https://x.com/deepseek_ai/status/1892786555494019098
“GPQA: 448 multiple choice questions in 16 subdomains SuperGPQA: 26,529 mutiple choice questions across 285 graduate disciplines 😲 DeepSeek-R1 outperforms o1, o2-mini, Claude 3.5 Sonnet, etc. on this benchmark 🤔 https://x.com/iScienceLuvr/status/1892879645223375319
“LLMs are still incredibly bad at long context, severe drop in response quality from the best of the best models (o1, Claude, grok, DeepSeek), doesn’t really matter what model – it will choke” / X https://x.com/abacaj/status/1893024046469493212
“📱 Turn any text into a podcast instantly! Transform articles, papers & blogs into audio content using open-source AI (deepseek-r1) + kokoro TTS. Think NotebookLM but fully open source 🎧 Nice work @ngxson! https://x.com/fdaudens/status/1891690883176604053
“Learn everything there is to know about LlamaParse in this comprehensive video! In this video, @mesudarshan covers: ➡️ Multiple parsing modes ➡️ Using parsing instructions to improve quality ➡️ Output formats available ➡️ Parsing audio and images ➡️ JSON mode ➡️ Using it all in https://x.com/llama_index/status/1890499579214491967
“Deepseek R1 just became the most liked model ever on @huggingface just a few weeks after release – with thousands of variants downloaded over 10 million times now! https://x.com/ClementDelangue/status/1890461279283769742
“Large Language Diffusion Models Introduces LLaDA-8B, a large language diffusion model that pretrained on 2.3 trillion tokens using 0.13 million H800 GPU hours, followed by SFT on 4.5 million pairs. LLaDA 8B surpasses Llama-2 7B on nearly all 15 standard zero/few-shot learning https://x.com/iScienceLuvr/status/1891337383200903625
“Our CEO @vipulved on Bloomberg @technology discussing Together AI’s $305M Series B announcement and why businesses are choosing open source AI. 🎥 “Enterprise leaders like @Zoom, @salesforce, and @SKtelecom are deploying open source AI with Together AI to maintain control over https://x.com/togethercompute/status/1892723230588514373
“🐋 DeepSeek-R1 Price Drop 🐋 Our serverless API for DeepSeek-R1 now has new, lower pricing: ⭐ $3.00 per million input tokens ⭐ $7.00 per million output tokens https://x.com/togethercompute/status/1892707292614709596
“Llamba: Scaling Distilled Recurrent Models for Efficient Language Processing “We introduce Llamba, a family of efficient recurrent language models distilled from Llama-3.x into the Mamba architecture. The series includes Llamba-1B, Llamba-3B, and Llamba-8B, which achieve higher https://x.com/iScienceLuvr/status/1892875837772615839
“LETS GOOOO – Hugging Face just released Ultra Scale Playbook for Training LLMs on GPU Clusters! 🤯 A free, open-source, book to learn everything about 5D parallelism, ZeRO, fast CUDA kernels, how and why overlap compute & communication – all scaling bottlenecks and tools https://x.com/reach_vb/status/1892276287039033473
“A team at @deepseek_ai plans to open-source 5 repositories next week, one per day. Focused on infrastructure and building blocks of their online services. https://x.com/_philschmid/status/1892857906669715779
ChatGPT comes to 500,000 new users in OpenAI’s largest AI education deal yet – Ars Technica https://arstechnica.com/ai/2025/02/chatgpt-comes-to-500000-new-users-in-openais-largest-ai-education-deal-yet/
LlamaCloud EU: Secure, compliant knowledge management for European enterprises 🇪🇺🔒 We’re excited to announce early access to LlamaCloud EU, our new SaaS offering that ensures full data residency within EU jurisdiction. This addresses a key barrier for European companies looking https://x.com/llama_index/status/1892271183451869512
“Announcing our first open-weights model: R1 1776 – a version of DeepSeek R1 that’s been post-trained to remove the China censorship and provide unbiased, accurate responses. Here’s a graph showing % of Chinese censorship by the model (the lower, the better). https://x.com/AravSrinivas/status/1891917148869755058
“Introducing @MistralAI batch API UI! Now you can easily create and monitor batch jobs right from la Plateforme: https://x.com/sophiamyang/status/1891869154770026502
“HuggingFace’s datasets and models platform is one of the most inclusive, backed by VCs. It hosts a wide range of content—academic and adult, copyright-free and copyrighted, liberal and conservative—attracting users not only from the West but also from the Chinese AI community.” / X https://x.com/arankomatsuzaki/status/1892971292003115032
“Let’s goo! Quite psyched to onboard @nebiusaistudio @novita_labs & @hyperbolic_labs over to Hugging Face Hub Inference Providers! 🔥 Starting today you can access SoTA VLMs (Qwen 72B VLM), Text to Image models like Flux AND LLMs like DeepSeek R1 with even more providers directly https://x.com/reach_vb/status/1891909914412433839
“Amazing new open-source text-to-video model just dropped. Image-to-video generation models struggle with synchronizing motion and maintaining realism. Existing diffusion-based models generate high-quality frames but fail in complex physics and logical sequences. Step-Video-T2V https://x.com/rohanpaul_ai/status/1891413622867288100
“April 29, see you at the first-ever LlamaCon. 🦙 https://x.com/AIatMeta/status/1891969855043313945
“@deepseek_ai Let’s go: https://x.com/_akhaliq/status/1892801072713908404
“Baichuan-M1 – Opensources SotA medical LLM (Baichuan-M1-14B) – Trained from scratch on 20T tokens with a dedicated focus on enhancing medical capabilities https://x.com/arankomatsuzaki/status/1892053102427300212
“R1-1776 has been released 🇺🇸. One more cool thing will be shipped this week. And one more cool thing will be announced next week. You’re going to like both. 🚢” / X https://x.com/AravSrinivas/status/1891958274880139484
“👀” / X https://x.com/fdaudens/status/1890410351507783941
“LlamaIndex.TS just got a lot smaller and easier to ship, check it out!” / X https://x.com/llama_index/status/1890502255683498139
“Introducing DeepHermes-3 Preview, a new LLM that unifies reasoning and intuitive language model capabilities. https://x.com/NousResearch/status/1890148000204485088
“🚀 We released the tech report of Qwen2.5VL ( https://x.com/Alibaba_Qwen/status/1892576737848160538
“FYI our discord has a community projects forum where you can find opensource projects to contribute to or start your own, join @NousResearch’s discord at https://x.com/Teknium1/status/1892747398914392197
“@noname874487118 Yep Liang Wenfeng did meet Xi Jinping. Bro is literally in the same suit and even the same pose on all meetings with Party elites. Looks stressful, hope he did alright https://x.com/teortaxesTex/status/1891365138529194473
“Let’s go back to a more open, transparent & collaborative AI (that’s IMO what fueled progress in 2016-2020)! https://x.com/ClementDelangue/status/1891590473153695860
“We are welcoming one of the most fan-favorite inference providers to @huggingface Hub, @FireworksAI_HQ 💜 https://x.com/mervenoyann/status/1890463154305405397
“@anish0209 Yes, Ollama is written in Go! Let’s go!” / X https://x.com/ollama/status/1891376667668426933
“@davidfowl Yes, we see downloads of models from Ollama going to the major cloud hosts.” / X https://x.com/ollama/status/1891581808451444932
ostris/Flex.1-alpha · Hugging Face https://huggingface.co/ostris/Flex.1-alpha
x.com/_akhaliq/status/1890215479047754194 https://x.com/_akhaliq/status/1890215479047754194
Red Hat’s take on open-source AI: Pragmatism over utopian dreams | ZDNET https://www.zdnet.com/article/red-hats-take-on-open-source-ai-pragmatism-over-utopian-dreams/
Mistral Saba | Mistral AI https://mistral.ai/news/mistral-saba
(4) An AI Alchemist and His DeepSeek Journey https://craftedminds.substack.com/p/an-ai-alchemist-and-his-deepseek
“The radical transparency here is incredible. Nobody is doing this at the level of DeepSeek” / X https://x.com/casper_hansen_/status/1892835887446159409
“Step-Video-T2V Technical Report The Practice, Challenges, and Future of Video Foundation Model 30B open-source text-to-video generation model https://x.com/_akhaliq/status/1891353872809025961
“Love this application: 📊 Codebase Analytics Dashboard 📊 Input an open source repo => it computes and visualizes health metrics Fully open source and built with @codegen . Build + ship your own 🚀 Here’s how 👇 https://x.com/mathemagic1an/status/1890531998063886829
“I feel like people slept on our @chai_research ep they straight up beat @character_ai at their own game*. like deepseek, former hedge fund guys going into the consumer LLM game mostly bootstrapped and crushing very hard. they just released their 2025 roadmap and… wow 25% https://x.com/swyx/status/1890475865680617982
“I am transitioning to focusing on my open source work full time. Expect more, a lot more, in the weeks to come. Open source models, ai-toolkit improvements / gui, tutorials, videos, and shamelessly asking for financial support. Speaking of which, Patreon link in profile.” / X https://x.com/ostrisai/status/1891820293993398609
“Grok 3 release with live demo on Monday night at 8pm PT. Smartest AI on Earth.” / X https://x.com/elonmusk/status/1890958798841389499
“Fully open-source 40B model for genome modeling and design across all domains of life by my colleague @MichaelPoli6 and team It also comes with a new StripedHyena 2 architecture.” / X https://x.com/maximelabonne/status/1892256596136165535
“Efficient Triton based implementation for DeepSeek’s Native Sparse Attention! 🔥 https://x.com/reach_vb/status/1893021796577714346
“AG2 v0.7.4 is here, packed with goodies! 🎁 🧠 Open Source eep Research with a model of your choice! 💬 Stay connected via Slack, discord, or Telegram with AG2’s Communication Agents Suite! Read full release note here: https://x.com/qingyun_wu/status/1890143734182080634
“🤯 Mind-blown by OmniParser V2. Game-changer for agents. It turns any computer screen into structured data 60% faster than v1 and works with all major LLMs. Open-source MIT license https://x.com/fdaudens/status/1891669358830645449
“🏟️Announcing @MistralAI Saba, our first regional language model. – Mistral Saba is a 24B parameter model trained on meticulously curated datasets from across the Middle East and South Asia. – Mistral Saba supports Arabic and many Indian-origin languages, and is particularly https://x.com/sophiamyang/status/1891487141718376580
“🎉 Excited to see everyone’s enthusiasm for deploying DeepSeek-R1! Here are our recommended settings for the best experience: • No system prompt • Temperature: 0.6 • Official prompts for search & file upload: https://x.com/deepseek_ai/status/1890324295181824107
“⏩ We’re focused on making Together AI the best place to run DeepSeek-R1 – and so we continue our efforts to accelerate inference. As we make our service increasingly efficient, we’re pleased to pass our savings on to you. Try it today → https://x.com/togethercompute/status/1892707294808310265
“@MistralAI Available via API `mistral-saba-2502` on la Plateforme! We also provide custom training with @MistralAI applied AI offerings: https://x.com/sophiamyang/status/1891488607518462157
“Grok 3 reasoning beta achieved 96 on AIME and 85 on GPQA, which is on par with the full o3. https://x.com/arankomatsuzaki/status/1891708250199839167
“Grok 3 is a new best model in the world from the @xai team! Grok 3 ranks #1 on Chatbot Arena w/a big gap, and scores impressively on pretraining and reasoning evals. congrats to @elonmusk @ibab @jimmybajimmyba @Yuhu_ai_ looking forward to more partnership on grok4 & beyond 🚀 https://x.com/alexandr_wang/status/1891714169629524126
“BREAKING: xAI announces Grok 3 Here is everything you need to know: https://x.com/omarsar0/status/1891705029083512934
“i used up my whole grok quota on “glub”. this fucking sucks https://x.com/andersonbcdefg/status/1892828532515991786
“i love the janitor, but just accept that Grok-3 is the most powerful PUBLICLY AVAILABLE LLM (at least for a day lol) Look at the condition of the bet. Grok-3 delivered. https://x.com/scaling01/status/1891842735834808708
“Grok 3 involved 10x more training than Grok 2! Grok finished pretraining in early January! The model is still training. https://x.com/omarsar0/status/1891705957220016403
“Elon mentioned that Grok 3 is an order of magnitude more capable than Grok 2. https://x.com/omarsar0/status/1891705031243469270
“Another question that might be answered by Grok 3 – there is some evidence that the impact of AI on productivity is driven by model scale as well, will it continue? (There are lots of reasons, though, that Grok 3 may not be as useful for work as other AI models, so we shall see)” / X https://x.com/emollick/status/1891359072546385995
“BREAKING: @xAI early version of Grok-3 (codename “chocolate”) is now #1 in Arena! 🏆 Grok-3 is: – First-ever model to break 1400 score! – #1 across all categories, a milestone that keeps getting harder to achieve Huge congratulations to @xAI on this milestone! View thread 🧵 https://x.com/lmarena_ai/status/1891706264800936307
“Grok-3 without reasoning actually looks pretty good on these 3 cherry picked benchmarks. It’s also a good sign that they got 1400 Elo n lmsys from the get go. However, I feel like this launch was rather underwhelming. Too few benchmarks, no report and no useful demos. If it’s https://x.com/scaling01/status/1891786871304323280
“Grok 3 on X Premium+ https://x.com/omarsar0/status/1891715441292083572
“Grok 3 release with live demo on Monday night at 8pm PT. Smartest AI on Earth.” / X https://x.com/elonmusk/status/1890958798841389499
“a man died to tell us how good grok 3 really is never forget https://x.com/aidan_mclau/status/1891243031090626776
Elon Musk’s Grok 3: Performance, How to Access, and More https://www.analyticsvidhya.com/blog/2025/02/grok-3/
“This is it: The world’s smartest AI, Grok 3, now available for free (until our servers melt). Try Grok 3 now: https://x.com/xai/status/1892400129719611567
Elon Musk says xAI’s Grok 3 chatbot to be unveiled on Monday | Reuters https://www.reuters.com/technology/artificial-intelligence/elon-musk-says-xais-grok-3-chatbot-be-unveiled-monday-2025-02-16/
“Grok 3 https://x.com/emollick/status/1891751665948045758
“Grok 3 Reasoning Beta performance on AIME 2025. Grok 3 shows generalization capabilities. It not only does coding and math problem-solving, but it can also do other creative and useful real-world tasks. https://x.com/omarsar0/status/1891711110476111884
“Grok 3 performance on AIME 2025 (math competition that just finished a few days ago!) https://x.com/iScienceLuvr/status/1891708408832610548
“I was given early access to Grok 3 earlier today, making me I think one of the first few who could run a quick vibe check. Thinking ✅ First, Grok 3 clearly has an around state of the art thinking model (“Think” button) and did great out of the box on my Settler’s of Catan https://x.com/karpathy/status/1891720635363254772
“Another thing Grok 3 highlights is the urgent need for better batteries of tests and independent testing authorities. Public benchmarks are both “meh” and saturated, leaving a lot of AI testing to be like food reviews, based on taste. If AI is critical to to work, we need more.” / X https://x.com/emollick/status/1891862982558187560
Grok 3, xAI’s New Model Family, Improves on its Predecessors, Adds Reasoning https://www.deeplearning.ai/the-batch/grok-3-xais-new-model-family-improves-on-its-predecessors-adds-reasoning/
“Reasoning models like Grok-3 reasoning beta and DeepSeek-R1 are trained using reinforcement learning with verifiable rewards, but what exactly does this mean? Verifiable tasks. One detail that we should immediately notice about reasoning models is that they are primarily used https://x.com/cwolferesearch/status/1891893034956030242
“Grok-3 should be open-sourced. @elonmusk @xai” / X https://x.com/huybery/status/1891712667947057598
“Grok 3 also has reasoning capabilities too! The Grok team has been testing these capabilities which they have unlocked using RL. The model is good, especially in coding. https://x.com/omarsar0/status/1891707915351859547
Update README.md · deepseek-ai/DeepSeek-R1@7ca5e1e https://github.com/deepseek-ai/DeepSeek-R1/commit/7ca5e1e7f75e12a1c561fffaa6aa686708f881ae
Downloads of DeepSeek’s AI apps paused in South Korea over privacy concerns | AP News https://apnews.com/article/south-korea-deepseek-app-downloads-privacy-concerns-ai-20950f357276b9bb8f2a70a4b3c03e96
Add fileupload and websearch prompt by DeepSeekPH · Pull Request #399 · deepseek-ai/DeepSeek-R1 https://github.com/deepseek-ai/DeepSeek-R1/pull/399/files
(4) An AI Alchemist and His DeepSeek Journey https://craftedminds.substack.com/p/an-ai-alchemist-and-his-deepseek
“Today we’re open-sourcing R1 1776—a version of the DeepSeek R1 model that has been post-trained to provide uncensored, unbiased, and factual information. https://x.com/perplexity_ai/status/1891916573713236248
“The radical transparency here is incredible. Nobody is doing this at the level of DeepSeek” / X https://x.com/casper_hansen_/status/1892835887446159409
“Serve DeepSeek-R1 671B on your infra! Serving a 671B model is challenging: 1. H100/H200s are scarce and expensive. 2. Multi-node inference is complex. SkyPilot eases cheap GPU hunting&managing and @lmsysorg’s SGLang handles the inference complexity: https://x.com/skypilot_org/status/1890449454840365110
“I will draw DeepSeek’s Native Sparse Attention in tomorrow’s live webinar ~ Attend 👉 https://x.com/ProfTomYeh/status/1892245518580932812
“DeepSeek releases any paper and the whole ML community stops to pay attention. That’s soft power my friend, and it’s gonna take them far.” / X https://x.com/hkproj/status/1892107497369915535
“btw, DeepSearch is also open source, you can find the code here: https://x.com/JinaAI_/status/1890410013031575574
“@deepseek_ai Nice – and thank you for your dedication to open source & science, don’t forget to claim the paper on Hugging Face: https://x.com/reach_vb/status/1891755094330212552
“Perplexity just released POST TRAINED DeepSeek R1 for factual and unbiased information – MIT Licensed 🔥 https://x.com/reach_vb/status/1891922768892989559
“Perplexity just dropped R1 1776 a version of the DeepSeek R1 model that has been post-trained to provide uncensored, unbiased, and factual information https://x.com/_akhaliq/status/1891961543455031429
“🎯 @perplexity_ai drops their FIRST open-weight model on @huggingface: A decensored DeepSeek-R1 with full reasoning capabilities. Tested on 1000+ examples for unbiased responses. https://x.com/fdaudens/status/1891949269470351833
“@deepseek_ai omg 5 repos real openai” / X https://x.com/Yuchenj_UW/status/1892787346317463782
“Less is More for Reasoning (LIMO): a 32B model fine-tuned with 817 examples can beat o1-preview on math reasoning! 🤯 Do we really need o1’s huge RL procedure to see reasoning emerge? It seems not. Researchers from Shanghai Jiaotong University just demonstrated that carefully https://x.com/AymericRoucher/status/1891822202812760206
“Fireworks ai is now a supported Inference Provider on Hugging Face you can run serverless inference to the following models deepseek-ai/DeepSeek-R1 deepseek-ai/DeepSeek-V3 mistralai/Mistral-Small-24B-Instruct-2501 Qwen/Qwen2.5-Coder-32B-Instruct https://x.com/_akhaliq/status/1890469849861619753
“Everyone vote for o3-mini type model to be open-sourced please 🥺🥺🥺 We can distill or quantize a phone sized model dw the open-source community will work its magic!!” / X https://x.com/iScienceLuvr/status/1891669417332805739
“joke aside it would be great to have o3-mini so it would be community distilling it instead” / X https://x.com/mervenoyann/status/1891772390301941796
“It’s as good as the o3-mini (openai’s reasoning model): https://x.com/Yuchenj_UW/status/1891732255938077133
“DeepSeek R1 671B just broke speed records at 198 t/s it is now the fastest reasoning model available you can try it in coding mode on anychat soon prompt: make two fancy D3.js animated looping speedometers, first one is title deepseek R1 671B and second one is OpenAI o3-mini, https://x.com/_akhaliq/status/1890215479047754194
“for our next open source project, would it be more useful to do an o3-mini level model that is pretty small but still needs to run on GPUs, or the best phone-sized model we can do?” / X https://x.com/sama/status/1891667332105109653
“> DeepSeek-R1 achieved the highest accuracy of 61.82% on SuperGPQA ByteDance Research btw, they have little reason to overhype it. I wonder how Grok does. It certainly has o3-mini and R1 beat on normal GPQA. https://x.com/teortaxesTex/status/1892849709053583386
The Ultra-Scale Playbook – a Hugging Face Space by nanotron https://huggingface.co/spaces/nanotron/ultrascale-playbook
facebook/natural_reasoning · Datasets at Hugging Face https://huggingface.co/datasets/facebook/natural_reasoning
