Image created with gemini-2.5-flash-image with claude-sonnet-4-5. Image prompt: An elegant overhead view of a spherical chess board designed as a world map, with continents forming alternating light and dark squares, ornate chess pieces bearing national emblems and cultural symbols positioned across different countries, dramatic side lighting casting long shadows across oceans, photorealistic with a slightly surreal quality, rich jewel tones and metallic finishes on the pieces.
Airbnb CEO Brian Chesky: “We’re relying a lot on Alibaba’s Qwen model. It’s very good. It’s also fast and cheap… We use OpenAI’s latest models, but we typically don’t use them that much in production because there are faster and cheaper models.” The valley is built on Qwen?”” / X https://x.com/natolambert/status/1980657338726887662
It makes perfect sense to let agents understand, imitate, and learn how humans use computers from videos! We present VideoAgentTrek, which builds strong computer-use agents through video pretraining and agentic tuning. This approach has already proven effective in the training of”” / X https://x.com/huybery/status/1981728838024560669
This AI trading benchmark is interesting. Each model got $10,000 to invest. ~3 days in: ranking atm: – DeepSeek V3.1: +$2,658 – Grok 4: +$2,236 – Claude 4.5 Sonnet: +$1,911 – Qwen 3 Max: −$211 – GPT-5: −$3,139 – Gemini 2.5 Pro: −$3,719 DeepSeek beats all the other models https://x.com/Yuchenj_UW/status/1980318499185823760
Qwen Deep Research just got a major upgrade. ⚡️ It now creates not only the report, but also a live webpage 🌐 and a podcast 🎙️ – Powered by Qwen3-Coder, Qwen-Image, and Qwen3-TTS. Your insights, now visual and audible. ✨ 👉 https://x.com/Alibaba_Qwen/status/1980609551486624237
Alibaba Cloud claims to slash Nvidia GPU use by 82% with new pooling system | South China Morning Post https://www.scmp.com/business/article/3329450/alibaba-cloud-claims-slash-nvidia-gpu-use-82-new-pooling-system
Introducing Qwen3-VL-2B and Qwen3-VL-32B! From edge to cloud, these dense powerhouses deliver ultimate performance per GPU memory, packing the full capabilities of Qwen3-VL into compact and scalable forms. 🔥 Qwen3-VL-32B outperforms GPT-5 mini & Claude 4 Sonnet across STEM, https://x.com/Alibaba_Qwen/status/1980665932625383868
🚨 WebDev Arena: Top 15 Disrupted! 4 new models have been added to the WebDev leaderboard: 🔸 #4 Claude Sonnet 4.5 Thinking 32k by @AnthropicAI 🔸 #4 GLM 4.6 (the new #1 open model) by @Zai_org 🔸 #11 Qwen3 235B A22B Instruct (and #7 open model) by @Alibaba_Qwen 🔸 #14 Claude https://x.com/arena/status/1980367208300835328
Thanks for sharing the internal benchmarks, @rauchg ! We love to see it. 🔥”” / X https://x.com/Kimi_Moonshot/status/1980219115840385349
DeepSeek’s new 685B MoE model attends to only to the most relevant tokens, delivering 2–3× faster long-context inference and 6–7× cheaper processing than its V3.1 model. The new v3.2 model has MIT-licensed weights, costs $0.28/$0.028/$0.42 per 1M input/cached/output tokens via https://x.com/DeepLearningAI/status/1980846573681520824
Massively unexpected update from DeepSeek: a powerful, high-compression MoE OCR model. > In production, DeepSeek-OCR can generate 33 million pages of data per day for LLMs/VLMs using 20 nodes (x8 A100-40G). They want ALL the tokens. You’re welcome to have some too. https://x.com/teortaxesTex/status/1980160624140456370
DeepSeek released an OCR model today. Their motivation is really interesting: they want to use visual modality as an efficient compression medium for textual information, and use this to solve long-context challenges in LLMs. Of course, they are using it to get more training https://x.com/iScienceLuvr/status/1980247935700066468
Text or Pixels? It Takes Half: On the Token Efficiency of Visual Text Inputs in Multimodal LLMs Looks like this paper is also exploring the direction DeepSeek is interested in: representing text more efficiently as images, observing almost half reduction in number of tokens https://x.com/iScienceLuvr/status/1980942325573648703
> by storing the data representation natively as image tiles This must be obvious but just to clarify: DeepSeek does not propose to store *screenshots* of your chat logs. Pixel representation can be ephemeral; what is stored is still tokens, just not *language* tokens. https://x.com/teortaxesTex/status/1980453820632297900
Karpathy is undoubtedly vision pilled. And thanks to this casual DeepSeek drop — so will you.”” / X https://x.com/bilawalsidhu/status/1980598830916939880
DeepSeek https://github.com/deepseek-ai/
@mervenoyann The good perf of DeepSeek models matches with what we observe on PrediBench! https://x.com/AymericRoucher/status/1980196484617523445
Again, I will reiterate this: DeepSeek was literally built by chinas top quant firm, and chinas TOP quants.”” / X https://x.com/hamptonism/status/1980182896049811780
After DeepSeek-V3.2-Exp added TileLang & CUDA ops, many asked: what exactly is TileLang? 🤔 In his post “”TileLang: 80 lines of Python kernel code to reach 95% of FlashMLA’s performance””, developer & Zhihu contributor ryume gives a full breakdown of this new AI programming https://x.com/ZhihuFrontier/status/1980170674112188440
GLM-4.6 providers overview: we are benchmarking API endpoints offered by Baseten, GMI, Parasail, Novita, Deepinfra GLM-4.6 (Reasoning) from @Zai_org is one of the most intelligent open weights models, with intelligence close to GPT-OSS-120b (high), DeepSeek V3.2 Exp (Reasoning) https://x.com/ArtificialAnlys/status/1980777360724226282
For people thinking that DeepSeek-OCR is the first model to render text as images, the University of Copenhagen already did this in 2023 Paper is called “”Language Modelling with Pixels””. They trained a Masked AutoEncoder (MAE) by rendering text as images and masking patches https://x.com/NielsRogge/status/1980559120760791125
We’re seeing a lot of usage around DeepSeek’s new OCR model. Alex packaged it so you can deploy and test it yourself – prompts and sample images included.”” / X https://x.com/basetenco/status/1980924381217104338
DeepSeek-OCR looks impressive, but its core idea is not new. Input “Text” as “Image” — already explored by: LANGUAGE MODELING WITH PIXELS (Phillip et al., ICLR 2023) CLIPPO: Image-and-Language Understanding from Pixels Only (Michael et al. CVPR 2023) Pix2Struct: Screenshot https://x.com/awinyimgprocess/status/1980506449706119642
A more serious thread on the DeepSeek-OCR hype / serious misinterpretation going on. 1. On token reduction via representing text in images, researchers from Cambridge have previously shown that 500x prompt token compression is possible (ACL’25, Li, Su, and Collier). Without https://x.com/Kangwook_Lee/status/1980709454522744902
DeepSeek finally released a new model and paper. And because this DeepSeek-OCR release is a bit different from what everyone expected, and DeepSeek releases are generally a big deal, I wanted to do a brief explainer of what it is all about. In short, they explore how vision https://x.com/rasbt/status/1980642191950090585
I quite like the new DeepSeek-OCR paper. It’s a good OCR model (maybe a bit worse than dots), and yes data collection etc., but anyway it doesn’t matter. The more interesting part for me (esp as a computer vision at heart who is temporarily masquerading as a natural language”” / X https://x.com/karpathy/status/1980397031542989305
🚨 DeepSeek just did something wild. They built an OCR system that compresses long text into vision tokens literally turning paragraphs into pixels. Their model, DeepSeek-OCR, achieves 97% decoding precision at 10× compression and still manages 60% accuracy even at 20×. That https://x.com/godofprompt/status/1980233080213590326
Letsss gooo! DeepSeek just released a 3B OCR model on Hugging Face 🔥 Optimised to be token efficient AND scale ~200K+ pages/day on A100-40G Same arch as DeepSeek VL2 Use it with Transformers, vLLM and more 🤗 https://x.com/reach_vb/status/1980170192392270227
a bunch of OCR models released in past few weeks: ~ deepseek-ocr-3b ~ olmo-ocr-2-7b ~ chandra-ocr-8b ~ nanonets-ocr2-3b ~ paddleocr-vl-0.9B ~ qwen3-vl-dense/moe (general vlm) ~ dots.ocr-3b Will be dropping a detailed comparison soon”” / X https://x.com/HarveenChadha/status/1981055277408669934
NEW DeepSeek OCR model that outperforms dots ocr while prefilling 3x less tokens https://x.com/casper_hansen_/status/1980166248878203093
DeepSeek-OCR has some weird architectural choices for the LLM decoder: DeepSeek3B-MoE-A570M -> uses MHA, no MLA (not even GQA?) -> 2 shared experts (like DeepSeek V2, but V3 only has 1) -> quite low sparsity, activation ratio is 12.5%. For V3 it’s 3.52%, for V2 it’s 5% -> not https://x.com/eliebakouch/status/1980193125202083951
I think Glyph coming out on the same day a) corroborates the results of DeepSeek OCR b) confirms the “they had it lying around for a while” suspicion. Charitably, they learned of Zhipu’s project retracing their steps and sped up the release. Other possibilities are obvious too.”” / X https://x.com/teortaxesTex/status/1980642000006451348
deepseek-ai/DeepSeek-OCR: Contexts Optical Compression https://github.com/deepseek-ai/DeepSeek-OCR
what happened this week with OCR and VLMs? * deepseek-ocr * chandra-ocr * nanonets-ocr2 * paddleocr-vl * qwen3-vl (2B, 32B, Instruct and Thinking) * dots.ocr * olmOCR 2 (based on Qwen2.5-VL) * LightOnOCR (smallies) top 5 trending models on @huggingface are still OCR/VLM! https://x.com/MaziyarPanahi/status/1981421331053760775
DeepSeek-OCR Contexts Optical Compression https://x.com/_akhaliq/status/1980260630780162505
DeepSeek OCR dropped … but honestly, Glyph [1], released the same day, showed something more interesting: 3–4× context compression and infilling cost reduction, no performance hit on long-context QA and summarization, which is much less trivial than OCR in many cases. If that https://x.com/arankomatsuzaki/status/1980722682246398069
🚀 DeepSeek-OCR — the new frontier of OCR from @deepseek_ai , exploring optical context compression for LLMs, is running blazingly fast on vLLM ⚡ (~2500 tokens/s on A100-40G) — powered by vllm==0.8.5 for day-0 model support. 🧠 Compresses visual contexts up to 20× while keeping https://x.com/vllm_project/status/1980235518706401405
This was unexpected Minimax M2 is the best Chinese model rn going toe to toe with Sonnet 4.5 Time to bump them up to A or S tier https://x.com/zephyr_z9/status/1981695536987357382
What in the F is an AI factory? I had to investigate what the unelected @EU_Commission is talking about today So according to them, it’s some data centers (which they call supercomputers) in 6 different EU countries I checked out the most powerful one: Karolina, a Czech data https://x.com/levelsio/status/1981351393513615813
US chip plant investment to outpace China, Taiwan and South Korea from 2027 – Nikkei Asia https://asia.nikkei.com/business/tech/semiconductors/us-chip-plant-investment-to-outpace-china-taiwan-and-south-korea-from-2027
Introducing Manus 1.5 Faster, better quality results. Unlimited context. Build full-stack web apps, with real AI features, backends, user logins, custom domains and analytics. Run your biz, start your side hustle, or just have fun. If you can dream it, Manus can build it. https://x.com/ManusAI_HQ/status/1978854323774792135
Introducing Mistral AI Studio, the production AI platform. Mistral AI Studio enables builders to move from AI experimentation to production with a robust runtime for agents and deep observability across the AI lifecycle. More on our blog: https://x.com/MistralAI/status/1981752578951233989
Introducing our new tiny vision language model: LFM2-VL-3B 👀 > Expanded multilingual visual understanding: English, Japanese, French, Spanish, German, Italian, Portuguese, Arabic, Chinese, Korean > 51.8% on MM-IFEval (instruction following) > 71.4% on RealWorldQA (real-world https://x.com/LiquidAI_/status/1980985540196393211
I expect GLM-4.6-Air to make an improvement similar to Qwen-3 to Q3-2507 update, or maybe even the latest Qwen round. Will be the default model between 30B and 200B.”” / X https://x.com/teortaxesTex/status/1981702360981557624
Choose the “”:exacto”” version of open-source models in Cline automatically route to the best inference provider for models like GLM-4.6, Qwen3-Coder, and Kimi-K2. Provider quality varies wildly, meaning the same model can yield completely different results at different endpoints. https://x.com/cline/status/1981370535176286355
Over the last 24 hours, I have finetuned three Qwen3-VL models (2B, 4B, and 8B) on the CATmuS dataset on @huggingface . The first version of the models are now available on the Small Models for GLAM organization with @vanstriendaniel ! (Link below). These are designed to work https://x.com/wjb_mattingly/status/1981736776076026044
Qwen3-VL-2B-Instruct app is out on Hugging Face https://x.com/_akhaliq/status/1980690335220351063
we just updated the model comparison on our blog for you 🫡 added Chandra, OlmOCR-2, Qwen3-VL and their averaged OlmOCR score! https://x.com/mervenoyann/status/1981396054634615280
Kimi K2 is up to 5x faster and 50% more accurate ：）”” / X https://x.com/crystalsssup/status/1980147163629047854
Do AIs think differently in different languages? https://www.theargumentmag.com/p/do-ais-think-differently-in-different
Qwen just released Qwen3-VL on Hugging Face The most powerful vision-language model in the Qwen series, with comprehensive upgrades across text understanding, visual reasoning, and long context video analysis. From GUI operations to 1M context. https://x.com/HuggingPapers/status/1980809413045940553
