AI11 min read
Perplexity releases pplx-embed-v2-late with a 0.6B edge model and a 9B model scoring 92.4%
Perplexity launched two ColBERT-style embedding models on Hugging Face under an MIT license.
From MarkTechPost
Loading
90 days of the enterprise, pro and team plans free, no card needed. See the plans
Blog
Daily notes on new models, LLM releases, agent frameworks and AI research, written from the sources we follow and delivered as a newsletter every day.
Get the morning and evening issues
Every new post of the day, in one email. Confirmation required.
AI11 min read
Perplexity launched two ColBERT-style embedding models on Hugging Face under an MIT license.
From MarkTechPost
Agents8 min read
Google launched EmbeddingGemma 2, a compact open-weight model that maps text, images, and audio into one vector space. It runs locally on phones with minimal RAM and enables instant on-device semantic search.
From Google AI developers blog
Photo: Yan Krukau
Agents9 min read
Google launched EmbeddingGemma 2, a compact open model that handles text, code, images, video, and audio. It uses a single shared vector space to enable unified search across all media types.
From Google AI developers blog
AI15 min read
Mistral AI launched Mistral Large 4 with one trillion parameters. It is not yet an open-weight model but will release weights in three weeks.
From TechCrunch AI
AI13 min read
Mistral released a preview of Mistral Large 4, a one-trillion-parameter model trained in Europe.
From The Decoder
AI13 min read
Mistral launched Mistral Large 4, nicknamed Le Chonk. It is a 1 trillion-parameter model available for free.
From Wired AI
AI10 min read
Reflection released Beam, an open-weight model that matches GLM 5.2 and Qwen 3.8 on benchmarks while using three to four times less compute.
From The Decoder
Models10 min read
Mistral AI released a public preview of its largest model, Mistral Large 4. The model features over one trillion parameters and excels in cybersecurity and coding tasks.
From Mistral AI news
AI12 min read
A new model called JEPA-Anything works across physics, biology, and medicine. It predicts future states by splitting them into multiple partial parts.
From The Decoder
LLMs8 min read
Tiiuae released Falcon-Emirati-7B to handle Emirati Arabic dialect nuances. It scores 84.83% on the Alyah benchmark, beating larger competitors.
From Hugging Face blog
Agents11 min read
Reflection released Beam, a 501 billion parameter text-only model for coding and science. Apache 2.0 weights are available this month after training on 23.8 trillion tokens.
From Latent Space
Agents6 min read
Zhipu AI released GLM 5.3, a massive 753B-parameter model, on Amazon Bedrock today.
From AWS machine learning blog
AI11 min read
Together AI released a free CLI tool that routes coding agents like Claude Code to open models.
From MarkTechPost
AI11 min read
Reflection AI released Beam, a 501B parameter MoE model with 23B active parameters. It targets coding and agentic workloads but is not yet available for self-hosting.
From MarkTechPost
LLMs10 min read
Databricks launched a new SQL join to run massive batch vector searches directly in its runtime engine.
From Databricks blog
AI16 min read
OpenAI solved the Navier-Stokes problem using internal agents but faced accusations of unethical data use and rushed announcements.
From The Verge AI
AI16 min read
Reflection AI launched Beam, a 501-billion-parameter text-only model designed to match Chinese open models in reasoning while using significantly less compute.
From TechCrunch AI
AI7 min read
Reka AI released a preview of Rho-1, a single model that handles text, images, video, and robot control.
From The Decoder
AI10 min read
OpenAI will watermark ChatGPT output in the European Union but makes its global API watermark optional.
From The Decoder
Posts are drafted from public feeds by models OpenSmartRoute routes to - the same router, skill and metering customers use - and always link to the original source. Corrections: support.
Archive (100)
