Aleph Alpha released Kolibri, a 78.1B-parameter English-German Mixture-of-Experts model that activates only 3.46B parameters per token, with a 1M-token context and per-request reasoning effort. The Ap…
DeepSeek released official macOS and Windows desktop apps for DeepSeek Harness v0.2, the MIT-licensed agent harness, adding a plugin manager, file and code-change review, and scheduled Automation Task…
NVIDIA's IsaacTeleop framework converts XR hand tracking and motion-controller input into commands for simulated and real robots through a graph-based retargeting engine, according to a Marktechpost t…
Prime Intellect launched Prime Inference, a serverless and reserved-capacity serving platform for frontier open-source models that processed nearly a trillion tokens per day internally before public r…
Cloudflare released Clef and Clef-flash, its first models trained by the Workers AI team, as open-weight decision models under Apache 2.0 that return typed probabilities instead of free-form text. Cle…
Cohere released Embed 5, an embedding model family in two tiers — Embed 5 Pro and Embed 5 Fast — that share one embedding space so users can index with Pro and query with Fast, with both generally ava…
NVIDIA released Kumo Tabular, a family of tabular foundation models in Small, Medium, and Large sizes spanning roughly 28M to 215M parameters that predict new rows in a single forward pass with no tra…
Perplexity Research and turbopuffer released pplx-embed-v2-context-9b-preview, a contextual embedding model for RAG pipelines that embeds each chunk with the full document in view. The model is traine…
Google DeepMind announced Gemini 4 Argon, its first Gemini 4-generation frontier model, which generates up to 1M output tokens in a single response versus 64K on earlier Gemini models and launches at …
Perplexity released Photon, an in-house Rust-based retrieval and ranking engine that cut p99 retrieval and ranking latency from about 800 ms to about 65 ms and now handles all production traffic for i…
NVIDIA researchers, with MIT and the University of Oxford, introduced Physis-Lang, a self-evolving physical-language framework that ranks first on the Physics-IQ Verified leaderboard snapshot dated Se…
Liquid AI released d1, a decision model that returns calibrated probabilities across a fixed set of typed outcomes in a single API call with zero generated output tokens, available today as a hosted A…
OpenAI launched dots, persistent GPT-6 Astra-powered AI agents that each run on their own cloud computer and browser, at its DevDay event, rolling out to ChatGPT Pro and Business Premium users in elig…
Google Cloud AI Research, with UNC-Chapel Hill, Stanford and Washington University in St. Louis, released RRSI (Regularized Recursive Self-Improvement), an Apache 2.0 research framework that lets an L…
Alibaba's Qwen team released Qwen-Audio-3.1, a five-model audio stack led by the full-duplex speech model Qwen-Audio-3.1-Realtime, available as a managed API on QwenCloud over WebSocket with no open w…
Anthropic released Claude Sonnet 5.5, a closed-weights model available on the Claude Platform, AWS, Google Cloud, and Microsoft Azure at unchanged list pricing of $2 per million input tokens and $10 p…
Fireworks AI released Ember-1, a post-trained version of Moonshot AI's open-weight Kimi K3 that delivers Kimi K3's quality with about 40% fewer tokens, available only through the Fireworks serverless …
TypeSafe AI released Jev, its first System One model, which takes unstructured state and returns typed decisions with calibrated probabilities rather than chat or code output. Jev is built on three pr…
Google Research introduced an AI video co-director suite of four agentic frameworks that turns short clips into coherent, minutes-long stories by targeting identity drift and cascading errors in multi…
Google Research's MSEB (Massive Sound Embedding Benchmark), version 0.1.0, is structured in three layers — types (Sound, SoundEmbedding, Score, TaskMetadata), the MultiModalEncoder contract that user …