https://tpsreport.news/news/meta-muse-ai-agent-apple-watch-wear-os TPS: Tokens Per Second en 2026-11-10T15:36:35.347Z Meta brings its Muse AI agent to Apple Watch and Wear OS with voice calls, text and dictation Meta, Muse, Apple Watch, Wear OS, AI agent, consumer AI, wearables https://tpsreport.news/news/gemini-robotics-2-5-factory-trials-asimov-jailbreak TPS: Tokens Per Second en 2026-11-10T15:36:14.632Z Gemini Robotics 2.5 reaches 96% success in Mercedes-Benz factory kitting trials gemini-robotics, google-deepmind, robotics, vla, apptronik, mercedes-benz, asimov-jailbreak, ai-safety https://tpsreport.news/news/adobe-max-creative-cloud-mcp-firefly-ai-assistant-agents TPS: Tokens Per Second en 2026-11-10T14:22:28.632Z Adobe previews MCP access letting third-party AI agents edit Photoshop, Premiere files Adobe, Adobe Max, Firefly, MCP, AI agents, Creative Cloud, Premiere, Model Context Protocol https://tpsreport.news/news/adobe-firefly-ai-assistant-recurring-tasks-planning-mode TPS: Tokens Per Second en 2026-11-10T14:21:47.536Z Adobe adds planning mode and preview of always-on recurring-task agents to Firefly AI Assistant Adobe, Firefly, AI agents, Creative Cloud, Premiere, MCP, Adobe Max https://tpsreport.news/news/fable-5-2-remote-labor-index-41-percent TPS: Tokens Per Second en 2026-11-10T14:21:15.760Z Anthropic's Fable 5.2 tops Remote Labor Index at 41.3%, completing real freelance jobs at professional quality Fable 5.2, Anthropic, Remote Labor Index, Scale Labs, CAIS, AI agents, benchmark https://tpsreport.news/news/adobe-max-2026-premiere-generative-remove-firefly-updates TPS: Tokens Per Second en 2026-11-10T14:20:44.648Z Adobe adds Generative Remove beta to Premiere and expands object removal across Photoshop and Lightroom Adobe, Adobe Max, Firefly, Premiere, Photoshop, Lightroom, Generative Remove, generative AI https://tpsreport.news/news/boost-ai-presis-self-hosted-classification-model TPS: Tokens Per Second en 2026-11-10T11:50:47.888Z boost.ai launches Presis, a self-hosted language model for AI agent classification, claims 2.5x speed gain boost.ai, Presis, self-hosted, classification, AI agents, voice AI https://tpsreport.news/news/claude-fable-5-2-jagged-agi-mollick TPS: Tokens Per Second en 2026-11-10T11:20:56.021Z Ethan Mollick: Anthropic's Claude Fable 5.2 is "almost" AGI, with gaps in taste and self-directed judgment Anthropic, Claude, Fable 5.2, AGI, Ethan Mollick, Jagged AGI, analysis https://tpsreport.news/news/gemini-create-edit-pdfs-documents-handwritten-notes TPS: Tokens Per Second en 2026-11-10T10:05:48.129Z Gemini now creates and edits documents, PDFs and spreadsheets in chat, including from photos of handwritten notes Gemini, Google, PDF, document editing, handwriting, product update https://tpsreport.news/news/underdog-woof-40mb-cpu-text-to-speech TPS: Tokens Per Second en 2026-11-10T08:21:27.944Z Underdog Woof, a 40.6 MB two-voice TTS model, claims 1.42% word error rate on a CPU text-to-speech, TTS, on-device, Underdog Woof, Hugging Face, CPU inference, open weights https://tpsreport.news/news/lingbot-touch-robbyant-tactile-foundation-model-open-source TPS: Tokens Per Second en 2026-11-10T07:36:01.127Z Ant Group's Robbyant open-sources LingBot-Touch, a tactile model trained on nearly 5,000 hours of data LingBot-Touch, Robbyant, Ant Group, tactile sensing, robotics, embodied AI, open source, Apache-2.0 https://tpsreport.news/news/hyperflow-h3-flash-14b-video-audio-few-step TPS: Tokens Per Second en 2026-11-10T06:21:01.012Z Video Rebirth releases HyperFlow H3 Flash, a 14B few-step video and audio model compressed from MiniMax-H3 HyperFlow H3 Flash, Video Rebirth, MiniMax-H3, video generation, audio generation, GGUF, open weights, ComfyUI https://tpsreport.news/news/claude-fable-5-2-code-ownership-prompting TPS: Tokens Per Second en 2026-11-10T05:50:50.091Z Anthropic: Telling Claude Fable 5.2 the Codebase Is Its Own Raised Code Merging to About Half of Tries claude-fable-5.2, anthropic, prompt-engineering, coding, llms https://tpsreport.news/news/seedream-5-5-100-round-image-editing-layers-pricing TPS: Tokens Per Second en 2026-11-10T05:36:41.688Z ByteDance's Seedream 5.5 Holds Up Through 100 Edit Rounds in QbitAI Test; 2K Edit Costs About 0.28 Yuan Seedream 5.5, ByteDance, image generation, image editing, Doubao, Jimeng, GPT-Image 2.6, API pricing https://tpsreport.news/news/google-deepmind-alphaproof-code-signal-protocol-verification TPS: Tokens Per Second en 2026-11-10T05:21:29.147Z Google DeepMind's AlphaProof Code formally verifies key Signal protocol guarantees in four weeks Google DeepMind, AlphaProof Code, formal verification, Signal protocol, cryptography, agentic AI, security https://tpsreport.news/news/anthropic-claude-model-guide-fable-5-2-opus-5-5 TPS: Tokens Per Second en 2026-11-10T05:21:10.330Z Anthropic says to start with Opus 5.5 at medium effort, reserves Fable 5.2 for hardest tasks Anthropic, Claude, Fable 5.2, Opus 5.5, Sonnet 5.5, Haiku 5.5, Claude Code, model selection https://tpsreport.news/news/openai-dots-multiple-local-computers TPS: Tokens Per Second en 2026-11-10T04:20:44.353Z OpenAI's dots agent can now connect to multiple local computers and split tasks across them OpenAI, dots, AI agents, GPT-6 Astra, GPT-6.1 Terra, product update https://tpsreport.news/news/nex-n2-6-nex-agi-321b-agentic-262k-context TPS: Tokens Per Second en 2026-11-10T03:50:50.723Z Nex AGI releases Nex-N2.6, a 321B-parameter agentic model with 262K context at $0.12/$0.40 per 1M tokens Nex AGI, Nex-N2.6, agentic AI, GLM-5.3-Flash, OpenRouter, model release, computer use, coding https://tpsreport.news/news/modelbest-les-minicpm-v-4-7-35b-a3b-iphone-20-tokens-per-second TPS: Tokens Per Second en 2026-11-10T02:20:58.373Z ModelBest's LES runs 35B MoE MiniCPM-V 4.7 at over 20 tokens/s on iPhone 18 Pro Max, company says ModelBest, MiniCPM-V, on-device AI, mixture-of-experts, LES, llama.cpp, speculative decoding, quantization https://tpsreport.news/news/anthropic-fable-5-2-mythos-5-2-llm-anthropic-0-31 TPS: Tokens Per Second en 2026-11-10T00:35:56.944Z Anthropic releases Fable 5.2 and Mythos 5.2; llm-anthropic 0.31 adds Fable 5.2 support anthropic, claude, fable-5.2, mythos-5.2, llm-anthropic, reasoning, simon-willison https://tpsreport.news/news/google-antigravity-knowledge-work-interface TPS: Tokens Per Second en 2026-11-10T00:20:44.942Z Google Antigravity adds Knowledge Work interface for non-coding tasks, with the same agent underneath Google, Antigravity, AI agents, product update, knowledge work, MCP https://tpsreport.news/news/ollama-v0-40-16-ripple-preview-65b-agentic-coding TPS: Tokens Per Second en 2026-11-09T23:50:53.503Z Ollama v0.40.16 adds Ricursive Intelligence's 65B dense Ripple Preview coding model Ollama, Ripple Preview, Ricursive Intelligence, agentic coding, local models, Qwen3.8 Flash Next, llama.cpp https://tpsreport.news/news/claude-fable-5-2-github-copilot-generally-available TPS: Tokens Per Second en 2026-11-09T22:50:41.840Z Claude Fable 5.2 is now generally available in GitHub Copilot Claude Fable 5.2, Anthropic, GitHub Copilot, GitHub, changelog https://tpsreport.news/news/anthropic-python-sdk-v2-7-0-claude-fable-5-2-claude-mythos-5-2 TPS: Tokens Per Second en 2026-11-09T22:36:02.889Z Anthropic Python SDK v2.7.0 adds claude-fable-5-2 and claude-mythos-5-2 model IDs Anthropic, Python SDK, claude-fable-5-2, claude-mythos-5-2, Bedrock, Vertex, Admin API, changelog https://tpsreport.news/news/cisco-ai-defense-claude-mythos-red-teaming TPS: Tokens Per Second en 2026-11-09T21:20:48.337Z Cisco adds Anthropic's Claude Mythos to AI Defense for red teaming, with no open prompt access Cisco, Anthropic, Claude Mythos, AI Defense, red teaming, AI security, Cisco Live, Splunk https://tpsreport.news/news/claude-fable-5-2-mythos-5-2-amazon-bedrock TPS: Tokens Per Second en 2026-11-09T20:21:03.657Z Claude Fable 5.2 and Mythos 5.2 now available on Amazon Bedrock and Claude Platform on AWS Claude Fable 5.2, Claude Mythos 5.2, Anthropic, Amazon Bedrock, AWS, Enterprise Frontier Safeguards, data retention https://tpsreport.news/news/apple-intelligence-usage-packs-logic-pro-49-99 TPS: Tokens Per Second en 2026-11-09T19:22:21.239Z Apple adds $49.99 Intelligence Usage Packs to Logic Pro for extended Apple Intelligence access Apple, Apple Intelligence, Logic Pro, Intelligence Usage Packs, Pricing, Apple Creator Studio, OpenAI, Google Gemini https://tpsreport.news/news/claude-fable-5-2-mythos-5-2-pricing-benchmarks TPS: Tokens Per Second en 2026-11-09T19:21:27.023Z Anthropic releases Claude Fable 5.2 and Mythos 5.2 at $8/$40 per 1M tokens, with an optional fast mode Anthropic, Claude Fable 5.2, Claude Mythos 5.2, pricing, benchmarks, agentic coding, cybersecurity https://tpsreport.news/news/tencent-qwen3-8-27b-dspine-b16-speculative-decoding TPS: Tokens Per Second en 2026-11-09T19:06:59.416Z Tencent releases DSpine draft model for Qwen3.8-27B: 16-token blocks, up to 3.84× agent speedup over autoregressive deco Tencent, DSpine, Qwen3.8-27B, speculative decoding, SGLang, draft model, Apache 2.0, inference https://tpsreport.news/news/bytedance-seedream-5-5-100-round-pixel-stable-editing TPS: Tokens Per Second en 2026-11-09T19:05:42.398Z Bytedance Seedream 5.5 claims pixel-stable image editing across up to 100 rounds Seedream 5.5, Bytedance, image generation, image editing, RGBA, Doubao, Jimeng, Volcengine https://tpsreport.news/news/meta-muse-ember-lightweight-agent-model TPS: Tokens Per Second en 2026-11-09T18:21:39.804Z Meta releases Muse Ember, its lightest model, at 10 cents per million input tokens Meta, Muse Ember, Meta Superintelligence Lab, agents, multimodal, pricing, Muse Wonder https://tpsreport.news/news/google-ai-studio-maps-platform-app-development TPS: Tokens Per Second en 2026-11-09T17:35:50.220Z Google AI Studio adds Maps Platform app building, with 100 free demo calls per API daily Google, Google AI Studio, Google Maps Platform, Gemini, developer tools https://tpsreport.news/news/google-gemini-memory-summary-view-edit-delete TPS: Tokens Per Second en 2026-11-09T17:05:42.671Z Google Gemini adds 'Memory summary' to view, edit and delete what the app remembers about you Gemini, Google, memory, Personal Intelligence, privacy, product update https://tpsreport.news/news/gemini-4-argon-robotics-erqa-87-1 TPS: Tokens Per Second en 2026-11-09T16:05:54.281Z Google releases Gemini 4 Argon for robot control, reports 87% on ERQA without helper tools Gemini 4 Argon, Google, robotics, ERQA, MuJoCo, Real2Sim, vision-language-action, Gemini API https://tpsreport.news/news/baidu-ernie-evolve-rsi-ernie-5-5-flash TPS: Tokens Per Second en 2026-11-09T15:50:54.886Z Baidu discloses ERNIE Evolve, an RSI research system already used to optimize ERNIE 5.5 Flash inference Baidu, ERNIE, ERNIE Evolve, ERNIE 5.5 Flash, RSI, recursive self-improvement, Kunlunxin P800, inference optimization https://tpsreport.news/news/elevenagents-campaign-manager-outbound-campaigns TPS: Tokens Per Second en 2026-11-09T15:20:39.440Z ElevenLabs launches ElevenAgents Campaign Manager to run outbound contacts to a defined outcome ElevenLabs, ElevenAgents, Campaign Manager, voice agents, outbound calling, WhatsApp https://tpsreport.news/news/nvidia-gr00t-2-human-video-robot-learning TPS: Tokens Per Second en 2026-11-09T14:50:57.415Z NVIDIA GR00T 2 reaches 40% zero-shot success on 50 unseen robot tasks using human video NVIDIA, GR00T 2, robotics, Cosmos-Predict2.5, CoRL, human video, zero-shot https://tpsreport.news/news/google-pathways-general-availability-single-controller-training TPS: Tokens Per Second en 2026-11-09T14:35:56.756Z Google makes Pathways, the distributed training system it built for Gemini, generally available on Google Cloud Pathways, Google Cloud, distributed training, Gemini, TPU, SPMD, RL post-training https://tpsreport.news/news/tencent-hy4-770b-49b-active-native-vision-computer-use TPS: Tokens Per Second en 2026-11-09T09:36:30.440Z Tencent releases Hy4: 770B total parameters, 49B active, with native vision and computer use Tencent, Hy4, Hunyuan, open-source, multimodal, computer-use, agents, WorkBuddy https://tpsreport.news/news/qianwen-office-2-0-team-agent-collaboration TPS: Tokens Per Second en 2026-11-09T09:05:58.667Z Qianwen Office 2.0 adds "Collaboration" so teams and Agents work in one shared space Qianwen Office, Team Agent, enterprise AI, AI agents, DingTalk, workplace automation, product update https://tpsreport.news/news/neohorse-1-5-flash-preview-tokenrhythm-180b-agent-model TPS: Tokens Per Second en 2026-11-09T07:05:56.280Z TokenRhythm opens limited beta for NeoHorse-1.5-Flash Preview, a 180B agent model with 6B active NeoHorse, TokenRhythm, agent model, Qwen3.8-Flash, limited beta, OpenSquilla, software engineering, China AI https://tpsreport.news/news/lg-k-exaone-2-5-free-api-access TPS: Tokens Per Second en 2026-11-09T02:35:44.361Z LG AI Research opens free API access to 750B-parameter K-EXAONE 2.5 K-EXAONE 2.5, LG AI Research, EXAONE, API access, free API https://tpsreport.news/news/tencent-yao-shunyu-ai-biology-risk-hy4 TPS: Tokens Per Second en 2026-11-08T15:50:42.637Z Tencent AI chief warns models will soon pose biology risks, days after Hy4 launch Tencent, Hy4, AI safety, biosecurity, open models, Yao Shunyu, cybersecurity