Thank you @mervenoyann for the shout-out!

ServiceNow-AI/Apriel-1.5-15b-Thinker running on a single GPU using `transformers serve` 🔥 great to have some very nice reasoning models that can run locally! next step, trying it on mps 👀

👏 Congratulations to @ServiceNowRSRCH on introducing Apriel-1.5-15B-Thinker — a powerful new AI model that delivers frontier-level reasoning with a fraction of the compute. We’re proud that our Nemotron collection helped power its training .

𝐁𝐑𝐄𝐀𝐊𝐈𝐍𝐆: @ServiceNow released a 15B parameter AI model today. The model is the product of a partnership with Turing, which provided the training data. Breakdown below.

ServiceNow has released Apriel-v1.5-15B-Thinker, a 15B open weights reasoning model that leads our Small Models category (<40B parameters) 💼 Overview: Apriel-v1.5-15B-Thinker is a dense, 15B parameter open weights reasoning model. This is not the first model ServiceNow has

🎊 Congratulations to @ServiceNowRSRCH on introducing Apriel-1.5-15B-Thinker — their 15B-parameter model that matches DeepSeek-R1-0528, Mistral-medium-1.2 and Gemini Flash 2.5 on the Artificial Analysis Index (AAI 52) — delivering comparable results at fraction of the size (at

SLAM Labs presents Apriel-1.5-15B-Thinker 🚀 An open-weights multimodal reasoning model that hits frontier-level performance with just a fraction of the compute.

This is an interesting technical LLM report. This 15B model beats QwQ32B while using quite fewer tokens. Most interestingly, the authors heavily use model merging to combine the strengths of different checkpoints. 🔗 https://t.co/thoIqNEeBd

🎉 Our work “Variable Layerwise Quantization: A Simple and Effective Approach to Quantize LLMs” is accepted at #ACLFindings2025 📎 https://t.co/7fKAnZQIBr — Keep key layers high-precision, push others lower → compact LLMs w/ ~no accuracy loss — Simple LIM & ZD scores rank layers

🚨🤯 Today Jensen Huang announced SLAM Lab's newest model on the @HelloKnowledge stage: Apriel‑Nemotron‑15B‑Thinker 🚨 A lean, mean reasoning machine punching way above its weight class 👊 Built by SLAM × NVIDIA. Smaller models, bigger impact. 🧵👇

🚨 SLAM Labs presents Apriel-5B! And it lands right in the green zone 🚨 Speed ⚡ + Accuracy 📈 + Efficiency 💸 This model punches above its weight, beating bigger LLMs while training on a fraction of the compute. Built with Fast-LLM, our in-house training stack. 🧵👇

BigDocs: An Open and Permissively-Licensed Dataset for Training Multimodal Models on Document and Code Tasks abs: https://t.co/l6wHdrGAt5 project page: https://t.co/55UGlS3FLQ BigDocs-7.5M is a high-quality, open-access dataset comprising 7.5 million multimodal documents across

🌟🌟🌟 We just released BigDocs: An Open Multimodal Dataset — our latest work on scaling document understanding across diverse data types! 📄 👉 Dive into the details: https://t.co/KfOKZKARDS 🧠 or come see us at the #NeurIPS2024 RBFM workshop! #AI @ServiceNowRSRCH #bigdocs

We introduce LLM2Vec, a simple approach to transform any decoder-only LLM into a text encoder. We achieve SOTA performance on MTEB in the unsupervised and supervised category (among the models trained only on publicly available data). 🧵1/N Paper: https://t.co/1ARXK1SWwR

📢📢Excited to share our new work 🍛CurryDPO 1/2 🔴Systematically curates multiple preference pairs and trains upon them in a curriculum learning setup with DPO framework 🔴Achieves notable performance gains over vanilla DPO method on MTbench, Vicuna, WizardLM, and UltraFeedback