Skip to content
Thursday, 19 November 2026
Tenesys AI News
Subscribe

AI Models

New models and versions from OpenAI, Anthropic, Google, Meta, Mistral, xAI, DeepSeek, Qwen and others.

Models· Important· 🧪 Worth Testing

NVIDIA Releases Nemotron 3.5 Transcribe for Multilingual Audio Analysis

NVIDIA has introduced Nemotron 3.5 Transcribe, an open-weight, promptable model for speech-to-text, speaker diarization, and translation. It utilizes a hybrid Mamba-Transformer architecture to achieve industry-leading performance with only 4 billion active parameters.

Hugging Face
Models· Important· 🧪 Worth Testing

Optimizing GPU Infrastructure with CUDA 13.4 Memory Enforcement

NVIDIA released CUDA 13.4, adding hardware-transparent memory boundary enforcement for GPU sharing. Paired with the new open-source GPU Fractioning Operator and KAI Scheduler v0.18.0, teams can now define specific memory requests, burst limits, and compute modes directly in Kubernetes Pod annotations.

NVIDIA Developer
Models· Important· 🧪 Worth Testing

LiquidAI releases LFM3 multimodal models for edge devices

LiquidAI has launched LFM3, a new generation of small foundation models built on a shared 2B-parameter backbone. The series includes text, vision, audio, and omni models designed for high-efficiency on-device deployment.

Hugging Face
Models· Important· 🧪 Worth Testing

NVIDIA Introduces Persona-Based Fine-Tuning Workflow for Localized LLMs

NVIDIA has released 'Nemotron-Personas' datasets for France and Belgium, designed to improve LLM accuracy regarding regional procedures, demographics, and cultural nuances. The update provides a complete post-training workflow—including data generation, LoRA fine-tuning via NVIDIA NeMo, and evaluation—to help developers transition from generic models to domain-specific agents.

NVIDIA Developer
Models· Important· 🧪 Worth Testing

Google DeepMind unveils Genie 4 for real-time world simulation

Google DeepMind introduced Genie 4, a multimodal world model that enables real-time interactive simulations. It natively generates video, ambient audio, and responsive dialogue, utilizing Gemini 3.8 Flash to maintain physical consistency and handle multi-step reasoning.

Google DeepMind
Models· Important· 🧪 Worth Testing

Boosting Large-Scale MoE Model Throughput on NVIDIA Hardware

NVIDIA has documented a threefold optimization approach to accelerate MoE inference, tested on the 124-billion-parameter Ling-3.0-flash model. By utilizing specialized Humming kernels, FP8 precision for the LM head, and Multi-Token Prediction, developers can achieve a 42.5% throughput increase and a 24.5% reduction in latency.

NVIDIA Developer
Models· Important· 🧪 Worth Testing

Google updates A2UI with swappable inference formats and Express mode

Google has updated its A2UI generative UI protocol to include swappable inference formats, addressing latency and reliability issues found in the previous Direct JSON approach. A new format, A2UI Express, reduces token usage by up to 3x and latency by up to 3x by using a concise Python-like syntax.

Google Developers
Models· Important· 🧪 Worth Testing

Google releases Gemini Robotics Embodied Reasoning 2 for enterprise production

Google has launched Gemini Robotics Embodied Reasoning 2 (ER 2) in public preview on the Gemini platform. This model acts as a high-level cognitive layer for robotics, capable of planning, self-correcting tasks, and managing multi-robot systems via natural language instructions.

Google Cloud AI
Models· Important· 🧪 Worth Testing

Boa Compressor Brings Online Long-Context Compression to LLMs

Researchers have introduced the Boa Compressor, an online tool utilizing a Mamba state-space model to compress long-form context into shorter representations in linear time. By distilling knowledge from full-context LLM interactions, this method allows for efficient per-query inference without the massive latency overhead of transformer-based compression.

arXiv cs.CL
Models· Important· 🧪 Worth Testing

Google releases Gemini Omni for video generation

Google has introduced Gemini Omni, a new video-generation model capable of creating and editing clips from text and image prompts. The model is now the default for Google's video workflows, while the previous Veo model has been deprecated in the API.

Zapier
Models· Important· 🧪 Worth Testing

Reduce AI agent costs on Bedrock with new prompt caching

Amazon Bedrock now supports explicit and implicit prompt caching for OpenAI models, allowing you to store and reuse static prompt prefixes like tools, instructions, and conversation history. By avoiding the re-processing of identical tokens across turns, you can significantly lower input costs and improve responsiveness in complex agent workflows.

AWS Machine Learning
Models· Important· 🧪 Worth Testing

IBM Spyre accelerator gains native PyTorch and vLLM integration

IBM has updated its software ecosystem to treat the Spyre dataflow accelerator as a first-class PyTorch backend. By leveraging TorchDynamo and TorchInductor, developers can now run models on IBM Z, LinuxONE, and Power systems using standard PyTorch workflows and vLLM plugins.

IBM Research
Models· Important

Gemini API now enables request storage by default for new paid projects

Google has updated the Gemini API to automatically store requests (store=true) for all new projects on the paid tier. This change aligns standard content generation methods with the Interactions API, requiring developers to manually opt out if they prefer not to store request data.

Gemini API Changelog
Models· Important· 🧪 Worth Testing

Z.ai GLM 5.3 Flash model now available with 1M-token context

Z.ai's GLM 5.3 Flash, a multimodal 320B-parameter Mixture-of-Experts model with 18B active parameters, is now available through Mistral's platform. It offers a 1M-token context window and is built for developer agent workflows, visual coding, and complex tool-use tasks. Pricing is set at $0.15 per million input tokens, $0.03 per million cached input tokens, and $0.50 per million output tokens.

Mistral Changelog
Models· Important· 🧪 Worth Testing

Cohere releases Rerank 5 with passage-level retrieval and calibrated scores

Cohere launched Rerank 5, its newest enterprise reranking models, in Pro and Fast tiers. Rerank 5 Pro leads major public benchmarks including BRIGHT, ViDoRe V3, BEIR and nanoMTEB, while Rerank 5 Fast offers near-leading quality at lower cost and higher throughput. The release introduces Context Focus, which returns the most relevant text spans within documents rather than whole documents, and calibrated relevance scores that read as probabilities across models and domains.

Cohere
Models· Important· 🧪 Worth Testing

NVIDIA expands Nemotron with new retrieval, reranking and document-parsing models for AI agents

NVIDIA has added three new open, commercially licensed components to its Nemotron Retriever lineup: Embed models (8B and 1B, in BF16 and NVFP4) for searching text and images, a Rerank model (8B) that reorders retrieved evidence by relevance, and a compact 631M Parse model that converts document images into structured text. The goal is to feed AI agents with better, more relevant evidence so they need fewer search rounds and fewer inference tokens to answer a question. NVIDIA reports the new Embed and Rerank models top the ViDoRe v3.1 leaderboards for dense retrieval and reranking respectively.

Hugging Face
Models· Important· 🧪 Worth Testing

AWS and i14 ditch the KV cache with a 3B linear-attention model on Trainium2

AI research company i14 partnered with AWS to port its Laplace 3B model, a 24-layer architecture built on Gated Linear Attention (GLA) instead of standard transformer attention, onto AWS Trainium2 using native PyTorch (TorchNeuron Native). Rather than a KV cache that expands with every token, the model keeps a constant 29.5 MB recurrent state per sequence. On a single trn2.48xlarge instance, the team reached about 120,000 training tokens per second and served 11,544 output tokens per second across 1,280 simultaneous sequences through an unmodified vLLM engine.

AWS Machine Learning
Models· Important· 🧪 Worth Testing

OpenAI adds optional 12-hour prompt caching to the Responses API

OpenAI's Responses API now supports a 12-hour prompt cache option, keeping shared instructions, tools and context cached for longer so repeated requests hours apart can still benefit from cheaper cached-input pricing. Writing to this longer cache costs 2.5x the standard input token price. This is an optional setting alongside the existing shorter default cache.

OpenAI API Changelog
Models· Important

Google launches native PyTorch inference engines for vLLM and SGLang on TPU

Google announced new vLLM and SGLang inference engines built directly on PyTorch for TPU, developed with Inferact and RadixArk respectively. The engines run on a new TorchTPU layer that makes TPUs a native PyTorch device, replacing the previous JAX-based backend approach. Google reports a public benchmark result of 2.01 cents per million tokens in high-concurrency configurations via the SemiAnalysis InferenceX project, plus support for agentic workloads, long context, disaggregated serving and multimodal models.

Google Developers