XOutlier
Download .zipAdd to Chrome

Databricks AI Research

@DbrxMosaicAI55k followers

We remove the barriers to state-of-the-art generative AI model development and make data + AI available to all.

Open on X
Top outlier
148×
Best organic post
Peak views
622k
Single post
Bookmarks
3.1k
Across 88 organic posts
Last scan
Oct 16
Community cache
Posts 88 posts
#DatePost↓ OutlierViewsLikesRepostsBookmarks
1Jun 30, '23Introducing training LLMs with AMD hardware! MosaicML + PyTorch 2.0 + ROCm 5.4+ = LLM training out of the box with zero code changes. With MosaicML, the ML community has additional148×622k638139163
2May 5, '23📢 Introducing MPT: a new family of open-source commercially usable LLMs from @MosaicML. Trained on 1T tokens of text+code, MPT models match and - in many ways - surpass LLaMa-7B. 86.5×363k1k206363
3Dec 15, '22Meet PubMed GPT 🩺 a new SOTA on the US Medical Licensing Exam developed by MosaicML and @StanfordHAI. It's a normal GPT-3B model trained on medical data that bests hand-designed m80.8×339k499129158
4Jun 26, '23Today, we’re excited to share that MosaicML has agreed to join @Databricks!75.9×319k6008535
5Jun 22, '23Meet MPT-30B, the latest member of @MosaicML's family of open-source, commercially usable models. It's trained on 1T tokens with up to 8k context (even more w/ALiBi) on A100s and *70.5×296k512115125
6May 3, '23📣Announcing MosaicML Inference 📣 Ever wanted a text or image generation API that doesn’t make you send data to a third party? Or a cheaper solution than paying by the token? Or a57.6×242k61889177
7Jul 19, '23📢 Today, we're thrilled to announce that @Databricks has completed its acquisition of MosaicML. Our teams share a common goal to make #GenerativeAI accessible for all. We're excit32.9×138k126229
8Apr 27, '23How good are @nvidia H100s actually? In collaboration with @CoreWeave, we benchmarked A100 vs H100 performance for large language model training. Here's what we found: [1/6]25.4×107k2124677
9Apr 28, '23[1/8] Full technical details on our Stable Diffusion 2.0 speedrun are here! On Wednesday, we announced that we had replicated SD2 for < $50k, 2.7x over our baseline and 6x over Sta16.5×69k3773864
10Jan 24, '23How much does it take to train a Stable Diffusion model from scratch? The answer: 79,000 A100-hours in 13 days, for a total training cost of &lt;$160k. Our tooling reduces the time16.1×67k1152539
11Oct 30, '23Our NLP architect @abhi_venigalla continues his work on the use of AMD accelerators at scale for #LLM training. In our latest @databricks blog post, he shares multi-node training p16×67k1102619
12Aug 2, '23How can the ML community measure LLM quality in a holistic and standardized manner? The Mosaic Model Gauntlet encompasses 34 benchmarks, organized into 6 broad categories of compet14.4×60k1613981
13Jul 18, '23Announcing MPT-7B-8K: a 7B parameter open-source LLM with 8k context length trained with the MosaicML platform. With its 8k context length, MPT-7B-8K specializes in document summar12.8×54k3347070
14Apr 12, '24Ready to use a programmatic approach to prompting #LLMs and building #RAG applications? The @stanfordnlp #dspy repo includes support for @databricks Model Serving and Vector Search12.6×53k1303388
15May 16, '24Excited to announce our #CVPR2024 paper, "CommonCanvas," featuring open diffusion models trained on Creative-Commons-licensed 📸 images! #AI #MachineLearning10.8×45k631329
16Aug 24, '23New Blog Post 📢 Llama2-70B-Chat is now available on MosaicML Inference and @databricks MLflow AI Gateway. Learn more:10.4×44k78209
18Apr 11, '23Woo hoo! 🙌What an honor to make the @Forbes AI 50 List. MosaicML empowers you build your own #GenerativeAI. Train, finetune, and deploy your custom #LLM today:10.3×43k314317
20Apr 5, '23Want to train your own custom #LLMs while keeping your data private? 🔒 We got you. The MosaicML platform keeps your data, models, and source code secure in your private network wh8.9×37k222228
21Mar 27, '24DBRX is a new open source general-purpose #LLM that advances the state of the art in efficiency, using a 132B-parameter MoE architecture. Check out our deep-dive on how we trained 8.6×36k1903844
22Sep 29, '22We have exciting news! In our latest and greatest LLM blog, we show how MosaicML Cloud can help you train LLMs from 1B - 70B parameters, and for the first time, publish transparent8.6×03274385
23Mar 10, '23📢 Introducing MosaicBERT! Now you can pretrain a high-quality BERT model from scratch on the MosaicML platform for $20. So why should you train your own BERT model? 👇 (1/5)8.4×35k94944
24Jun 9, '23We launched the MPT-7B foundation models just over a month ago, and since then, they’ve been downloaded over 2 million times! We are humbled by this warm reception, and thrilled to8.2×34k79135
25Feb 7, '23There's fast, and then there's blazingly fast. 🔥🔥🔥 With Composer and MosaicML Cloud, you can now evaluate #llms on in-context learning tasks (LAMBADA, HellaSwag, PIQA, and more)8.0×33k62813
26Aug 11, '22Large Language Models (LLM) are gaining in popularity, but training these models from scratch can be a huge pain... until now! Our latest LLM blog series uncovers how to reduce the7.6×024139109
27Feb 28, '23Think it’s too hard—or too expensive—to train your own GPT or diffusion models from scratch? Think again. We built the MosaicML platform to tackle the challenges of training large 7.2×30k471023
28Apr 26, '23🚨 A few months ago we announced that you can train Stable Diffusion from scratch for less than $125k using the MosaicML platform. A major price drop is coming...and we have the tr6.6×28k7097
29Oct 18, '22📢 MosaicML Cloud is now available for early access! Create advanced AI models faster and cheaper than you thought possible.5.8×02784918
30Apr 25, '22Today, an exciting paper from @MSFTResearch: Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer While it's too early to say, this may be remember5.7×02053864
31Apr 4, '23Got an extra $20 burning a hole in your wallet? With the MosaicBERT architecture + training recipe, you can now pretrain a competitive #BERT-Base model from scratch on the MosaicML5.7×24k3081011
32Aug 12, '24How good do the latest long context LLMs (LLama-3.1-405b, GPT-4o-mini and Claude-3.5-sonnet) perform on RAG? We benchmarked 13 popular OSS and commercial models on context lengths 5.2×22k971861
33Oct 13, '21Hello World! Today we come out of stealth to make ML training more efficient with a mosaic of methods that modify training to improve speed, reduce cost, and boost quality. Read ou5.1×01594143
34Aug 8, '22"(Certified!!) Adversarial Robustness for Free!" Paper: This is a rare paper where having multiple exclamation marks in the title is justified. [1/12]5.1×01562579
35Mar 20, '24Since becoming part of @databricks last July, the MosaicML team has continued its mission to optimize and improve #GenAI model training. Our rigorous science leads to real-world re4.2×18k49720
36Feb 8, '22Is forgetting actually beneficial for training? This top-reviewed ICLR paper () introduces a powerful paradigm that consistently outperforms all other methods on ResNet18 and Dense4.2×01442945
37Feb 15, '24What should you do if you want to effectively and cheaply “instruction finetune” an LLM? @aditi_jh and @JacobianNeuro share some important insights. (1/5)4.2×18k51729
38Mar 4, '24New blog post! @zeqiuwu1, @huyushi98, and @rajammanabrolu share a recent highlight from their work in #LLM finetuning research: Fine-Grained Reinforcement Learning from Human Feedb4.2×18k611134
39May 8, '25We’re proud to be a platinum sponsor of #MLSys2025 alongside our co-founder & CTO @matei_zaharia serving as general chair. Stop by our booth to check out the latest projects from t4.2×17k33153
40Apr 18, '22When Shift Operation Meets Vision Transformer: An Extremely Simple Alternative to Attention Mechanism They replace the attention mechanism in a Swin transformer with a simple spati4.1×01282156
41Jul 19, '24Popular #LLM scaling laws only factor in training costs, and ignore the costs of deployment. In a paper presented at @icmlconf 2024, @databricks Mosaic AI researchers Nikhil Sardan4.0×17k531932
42Dec 14, '21Is overparametrization the key to solving ImageNet? In this NeurIPS Outstanding Paper Award winner, , @SebastienBubeck and @geoishard are looking at the significance of neural netw3.9×01492438
43Jul 5, '23Thanks @AMD for sharing our blog post about we trained #LLMs on your Instinct accelerators with no code changes. It just works!3.9×16k5381
44Apr 1, '24DBRX is the top open-source model on the latest WildBench Leaderboard on HuggingFace! Thanks to our friends @allen_ai for this benchmark of LLMs with challenging tasks from real us3.8×16k902912
45Mar 28, '24Our #DBRX open source #LLM is now available on ! Sign in (or sign up), select custom models, and start chatting with our DBRX-Instruct model. Get started:3.7×15k701516
47Aug 10, '23📦 To evaluate the coding capabilities of LLMs, you need to execute the code. But what if the LLM spits out malicious code?😱 With MosaicML, you can now evaluate #LLMs on code gen 3.5×15k541114
48Aug 16, '24Function calling significantly enhances the utility of LLMs in real-world applications; however, evaluating and improving this capability isn't easy — and no one benchmark tells th3.5×15k381135
49Apr 15, '22"Scaling Laws for Neural Language Models" was one of the papers that fueled the recent push towards larger models. @DeepMind revisits the question, "How should one scale model size3.4×01162636
50Jul 30, '24For our StreamingDataset users: We're thrilled to announce support for storing MDS datasets in @huggingface. S/O to @orionweller for the contribution! Check out the docs here:3.3×14k55128
51Jan 26, '22Data2Vec: A new paper by Meta AI claims to be “The first high-performance self-supervised algorithm that works for speech, vision, and text.” And the results look very promising.3.2×01201639
52Oct 10, '24New blog post! @quinn_leng, @JacobianNeuro, @sam_havens, @matei_zaharia, and @mcarbin have released part 2 of our investigation into long-context RAG! Read on for the full details:3.1×13k531824
53Mar 16, '22We've shared great research before, but reproducing methods from papers is hard. Announcing Composer, our library of ML speedups: . Train CV models ~4x faster and NLP models ~2x fa2.9×01023214
54Jul 25, '22"RankGen: Improving Text Generation with Large Ranking Models" It's tempting to look at this paper as yet another method to make the numbers go up. But there's another story here t2.9×0952132
55Jan 11, '22New blog post! Take a look at some best practices for efficient CNN training, and find out how you can apply them easily with our Composer library: #EfficientML2.8×0771840
56Nov 25, '24New blog post from @mansiege, @_BrettLarsen, @cojennin, and @code_star on Continued Pre-Training (CPT):2.8×12k699814
57Sep 19, '24At @databricks, we want to help customers build more #inference-friendly #llms. With MixAttention architecture, you can maintain model quality while improving inference speed and r2.7×11k651517
58Mar 2, '22Today we take a look at a new architecture from DeepMind called RETRO: Retrievel-Enhanced Transformer (). RETRO uses a large database of documents along with an embedding-based ret2.7×0891735
59Feb 1, '24We're excited to have contributed to the development of the OLMo open source model from @allen_ai which was developed using the @databricks Mosaic AI training platform. Read the bl2.7×11k59109
60Jan 17, '25We're kicking off 2025 with another Compound AI System Meetup with @lancedb in Mountain View on Jan 22! 🎉 Join us for a deep dive into AI infrastructure and insights with Lu Qiu, 2.7×11k36858
61Jun 6, '23Did somebody say "best-in-class open-source generative models?" Thanks for the recognition, @databricks! Check out our MPT-7B series here:2.7×11k59912
62Oct 13, '23Our #GenAI engineering team shares tips and tricks on how to deploy open source #LLMs for production usage in their latest @databricks blog post.2.7×11k631020
63Apr 11, '22We're excited to share details about how we use cyclic LR schedules to estimate time/cost/accuracy tradeoff curves for model training. This is a key element in our approach to #Eff2.7×0931531
64Jun 29, '22Today, we're excited to share our MLPerf Training 2.0 results: 23.8 minutes to train ResNet-50 to 75.9% validation accuracy on ImageNet, using 8x NVIDIA A100 GPUs.2.6×0961920
65Dec 2, '21Happy December! Today, we're looking back at Stochastic Weight Averaging (SWA), now a classic ML efficiency win! SWA is a simple method for improving accuracy with no increase in t2.6×0902224
67Dec 11, '24Come for the bricks, stay for the @databricks DSPy demo. #NeurIPS20242.3×9.8k58726
68Dec 19, '24New blog post on Benchmarking Domain Intelligence: Evaluating your #AI solutions should be done with tests that match your actual use case. We observed that the tasks in many acade2.3×9.7k499817
69Dec 4, '21Today, we're looking at fine-tuning large models, and this paper submitted to ICLR: It shows fine-tuning can hurt performance on out-of-distribution examples, and explains how usin2.3×0731529
71Feb 17, '23The MosaicML team is excited to present at the @wandb webinar this Thursday, 23-Feb-2023, 7PM CET/10AM PST! Our very own @leavittron will be joining W&amp;B's @carey_phelps to show2.1×8.9k209120
73Feb 4, '22Today, we're looking at a recent paper out of DeepMind outlining ReLICv2 (), which shows an exciting step forward in self-supervised learning! S/O to @weballergy2.1×0691722
76Aug 1, '22"ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers" Paper: Post-training quantization from the DeepSpeed team that can reduce inferences c1.9×067724
77Nov 17, '21Is this the end of contrastive self-supervised pretraining? In this :thread:, we’ll discuss Masked Autoencoders Are Scalable Vision Learners, the exciting new work from Kaiming He,1.9×0641321
78May 31, '23Join us next week at @wandb' Fully Connected conference on Wednesday, June 7th in San Francisco. Our CTO/Co-Founder @hanlintang will be speaking alongside a roster of #generativeAI1.9×7.9k3795
79Jul 11, '22"A Neural Corpus Indexer for Document Retrieval" Paper: They train a seq-to-seq model to directly spit out document IDs given queries. And it works really well. Like, these are som1.8×0661218
80Dec 11, '21Should you train your ViT? While Vision Transformers (ViT) have delivered groundbreaking performance, that performance often depends on tons of pre-training and data augmentations.1.8×066722
82Jan 14, '22Today, we look at optimizing data movement for transformer deep learning networks. In this paper (), Ivanov et al. show that 40% of the runtime is spent in data movement, and that 1.8×0481127
83Jul 14, '22We’ve released Composer 0.8.0, which introduces a HuggingFaceModel object for reading in your existing 🤗 Transformers models. Training or fine-tuning BERT models with Composer jus1.4×055148
84Feb 18, '22Do language models really need tokenizers? Work by @colinraffel's group suggests the answer is often no. Their Byt5 model () modifies mT5 to take in raw UTF-8 bytes instead of outp1.3×053711
85Jun 27, '22Today, we will take a look at the [Dataset Cartography]() paper that analyzes how sample classification changes during training.1.2×045910
86Mar 18, '22Composer is trending on GitHub (python)! Composer helps train ML models faster and cheaper through algorithmic efficiency, and the world is taking notice, thanks to this wonderful 1.2×0401211
87Jun 20, '22"Maximizing Communication Efficiency for Large-scale Training via 0/1 Adam" One of the biggest bottlenecks in distributed training is communication between nodes. The 0/1 Adam opti1.2×048810
89Nov 22, '21Large Language Models are notorious for being expensive to train, but provide a model that can be evaluated on generalized language understanding benchmarks. What if the goal is to1.1×042514
90Jun 16, '22Composer gives you world-class model accuracy at a fraction of the cost. FFCV is a super effective tool that helps us do it! Our latest blog post breaks it down: S/O to @gpoleclerc1.1×040114
91Feb 1, '22Today, we’re looking at “Augmenting Convolutional networks with attention-based aggregation” which proposes a modification to the average pooling used in many conv nets and a parti1.1×039513
92Dec 7, '21Does your application require high accuracy, but has tight inference constraints? Are you willing to pay any training costs to achieve this? If so, today’s paper may be for you! Th1.1×039512
93Mar 17, '22Shoutout to @jeremyphoward, @math_rachel and the whole @fastdotai team. Two-way callbacks and other ideas helped a lot in designing composer (). We're standing on the shoulders of 1.0×04875
94Dec 16, '24We're back from Vancouver and #neurips2024—thanks to all the #genai researchers, practitioners and international pop stars who joined us at our @databricks Mosaic AI social event!1.0×4.2k25783
97Jan 19, '22Exciting new speedups from @aleks_madry's lab with results on #Imagenet! Optimized data-loading with FFCV removes the CPU bottleneck which normally limits the throughput of ResNet+0.9×03893
98Dec 12, '24Last day of the #neurips2024 expo! Come by the @databricks booth for free toques, t-shirts, and locally-sourced, free-range bricks!0.9×3.7k40962
99Jul 18, '22“Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models” GLUE is worn out. Even SuperGLUE doesn’t stick anymore. Enter BIG-bench, a collection0.8×03745
100Dec 11, '24#TaylorSwift may have wrapped up the Eras Tour but we’re still in our Data and AI era! Stop by our booth at #NeurIPS2024 to chat all things research and meet #Brickster Swifties. F0.6×2.7k24940

Scan another creator

@
@karpathy@levelsio@paulg@naval@marclou@illyism

Run this on any account.

XOutlier builds this table for every public profile you open on x.com, with live data and every post, not just the top 50.

Add to Chrome Download .zip