https://tpsreport.news/news/meta-muse-spark-contributor-pricing-data TPS — Tokens Per Second en 2026-09-03T18:35:47.986Z Meta Offers 92-95% Discount on Muse Spark AI Model to Users Who Share Their Prompts Meta, Muse Spark, AI pricing, training data, coding agents, enterprise AI https://tpsreport.news/news/openai-astra-model-release TPS — Tokens Per Second en 2026-09-03T18:20:51.394Z OpenAI Releases Astra, Claims New Flagship Model Beats Rivals on Coding and Cybersecurity Benchmarks OpenAI, Astra, model release, AGI, AI alignment, cybersecurity, reasoning models, chain of thought https://tpsreport.news/news/google-weathernext-3-precipitation-forecasts TPS — Tokens Per Second en 2026-09-03T18:05:55.791Z Google Launches WeatherNext 3, Claims 50% More Accurate Precipitation Forecasts Google DeepMind, WeatherNext 3, weather forecasting, Gemini, Google Search, AI model release, climate AI https://tpsreport.news/news/ling-3-0-flash-fin-finance-model TPS — Tokens Per Second en 2026-09-03T16:50:50.170Z InclusionAI Releases Ling 3.0 Flash Fin, a Finance-Focused MoE Model with 5.1B Active Parameters InclusionAI, Ling 3.0, mixture-of-experts, finance AI, OpenRouter, model release https://tpsreport.news/news/github-copilot-app-parallel-agents-guide TPS — Tokens Per Second en 2026-09-03T16:06:33.733Z GitHub Copilot App Adds Guide for Running Multiple Agents in Parallel GitHub Copilot, AI agents, developer tools, product update https://tpsreport.news/news/github-copilot-app-parallel-agents-beginner-guide TPS — Tokens Per Second en 2026-09-03T16:05:54.197Z GitHub Publishes Beginner Guide to Running Multiple Copilot Agents in Parallel GitHub Copilot, coding agents, product update, developer tools, agentic workflows https://tpsreport.news/news/weathernext-3-google-deepmind-weather-model TPS — Tokens Per Second en 2026-09-03T15:20:55.931Z Google DeepMind Launches WeatherNext 3, Cutting Forecast Grid to 5km Using Live Satellite Data WeatherNext, Google DeepMind, weather forecasting, satellite data, climate AI, Google Cloud https://tpsreport.news/news/claude-fable-5-1-decodes-1653-royalist-cipher TPS — Tokens Per Second en 2026-09-03T14:35:49.570Z Anthropic's Claude Fable 5.1 Reportedly Solves 1653 Royalist Cipher in 44 Minutes Anthropic, Claude, cryptography, AI reasoning, Vals AI https://tpsreport.news/news/gemini-3-8-flash-coding-agentic-pricing TPS — Tokens Per Second en 2026-09-03T12:35:56.365Z Google Launches Gemini 3.8 Flash With Same Pricing as Predecessor, Its Third Flash Model in Six Weeks Gemini, Google DeepMind, Gemini 3.8 Flash, Alphabet, AI pricing, cybersecurity AI, enterprise AI https://tpsreport.news/news/meta-muse-spark-1-3-pricing-benchmarks TPS — Tokens Per Second en 2026-09-03T12:06:02.825Z Meta Releases Muse Spark 1.3, Cheapest Model in Its Performance Class at $0.55 Per Task Meta, Muse Spark, model release, pricing, benchmarks, agentic AI, Artificial Analysis https://tpsreport.news/news/muse-spark-1-3-meta-superintelligence-frontier TPS — Tokens Per Second en 2026-09-03T04:50:53.871Z Meta's Muse Spark 1.3 Claims #3 Global Ranking, Matches OpenAI's GPT-5.6-Sol on Coding Benchmarks Meta, Muse Spark, Meta Superintelligence Labs, GPT-5.6-Sol, OpenAI, model release, open weights, benchmarks https://tpsreport.news/news/openai-astra-opaque-recurrence-safety-concerns TPS — Tokens Per Second en 2026-09-02T20:35:47.239Z OpenAI's Reported 'Opaque Recurrence' Technique in Upcoming Astra Model Alarms AI Safety Researchers OpenAI, AI safety, chain-of-thought, reasoning models, Astra, Redwood Research https://tpsreport.news/news/gemini-3-8-flash-google-pricing-benchmarks TPS — Tokens Per Second en 2026-09-02T20:20:57.632Z Google Launches Gemini 3.8 Flash, Warns It May Use More Tokens Despite Unchanged Pricing Gemini 3.8 Flash, Google, Google DeepMind, pricing, benchmarks, AI agents, cybersecurity https://tpsreport.news/news/meta-muse-spark-1-3-contributor-release TPS — Tokens Per Second en 2026-09-02T19:51:35.616Z Meta Releases Muse Spark 1.3 Contributor, a Low-Cost Multimodal Reasoning Model With 1M Context Window Meta, Muse Spark, multimodal, reasoning model, OpenRouter, model release https://tpsreport.news/news/meta-muse-spark-1-3-1m-context TPS — Tokens Per Second en 2026-09-02T19:50:47.170Z Meta Releases Muse Spark 1.3, a Free Multimodal Reasoning Model with 1M-Token Context Meta, Muse Spark, multimodal, reasoning model, long context, agentic AI, OpenRouter https://tpsreport.news/news/claude-background-computer-use-mac TPS — Tokens Per Second en 2026-09-02T19:21:27.999Z Anthropic Brings Background Computer Use to Claude Code and Cowork on Mac Anthropic, Claude, Claude Code, Claude Cowork, computer use, macOS, product update https://tpsreport.news/news/github-copilot-cost-efficient-ai-coding TPS — Tokens Per Second en 2026-09-02T18:06:07.237Z GitHub Explains How Copilot Cuts AI Coding Costs Without Lowering Task Quality GitHub Copilot, AI coding assistants, cost efficiency, agentic coding, LLM engineering https://tpsreport.news/news/github-copilot-cost-efficiency-coding-tasks TPS — Tokens Per Second en 2026-09-02T18:05:42.502Z GitHub Says Shorter AI Outputs Don't Always Mean Lower Cost, Details Copilot Efficiency Approach GitHub Copilot, AI coding assistants, cost efficiency, developer tools, GitHub https://tpsreport.news/news/openai-astra-safety-concerns-opaque-reasoning TPS — Tokens Per Second en 2026-09-02T16:50:45.068Z Safety Researchers Warn OpenAI's Unreleased Astra Model May Hide Its Reasoning From Monitors OpenAI, AI safety, Astra, chain-of-thought, AI alignment, model architecture https://tpsreport.news/news/gemini-3-8-flash-and-cyber-variant TPS — Tokens Per Second en 2026-09-02T16:36:31.666Z Google DeepMind Ships Gemini 3.8 Flash and a Cybersecurity Variant, Third Flash Release in Six Weeks Gemini 3.8, Google DeepMind, Gemini Flash, AI cybersecurity, model release, Fairwind Program https://tpsreport.news/news/gemini-3-8-flash-rolling-out TPS — Tokens Per Second en 2026-09-02T15:36:28.857Z Google Rolls Out Gemini 3.8 Flash, Third Flash Update in Three Months Gemini, Google DeepMind, Gemini 3.8 Flash, Google AI Studio, model update, pricing https://tpsreport.news/news/gemini-3-8-flash-1m-context-openrouter TPS — Tokens Per Second en 2026-09-02T15:35:57.089Z Google Lists Gemini 3.8 Flash on OpenRouter With 1M-Token Context, September 2026 Release Date Gemini, Google, Gemini 3.8 Flash, OpenRouter, long context, Flash model https://tpsreport.news/news/us-military-adds-chatgpt-grok-genai-mil TPS — Tokens Per Second en 2026-09-02T14:51:03.292Z Pentagon Adds OpenAI's ChatGPT Mil and xAI's Grok for Government to GenAI.mil ChatGPT, Grok, Pentagon, GenAI.mil, government AI, OpenAI, xAI, Anthropic https://tpsreport.news/news/claude-system-prompt-song-lyrics-ban TPS — Tokens Per Second en 2026-09-02T14:36:34.258Z Anthropic Adds Explicit Song Lyric and Copyrighted Character Bans to Claude's System Prompt Anthropic, Claude, system prompt, copyright, song lyrics, AI policy, Fable 5.1 https://tpsreport.news/news/openai-astra-critical-cyber-risk-rating TPS — Tokens Per Second en 2026-09-02T14:36:02.807Z OpenAI Rates Upcoming Astra Model 'Critical' Risk for Cyber Capabilities — Its Highest Tier Ever OpenAI, Astra, cybersecurity, AI safety, chain-of-thought monitoring, Preparedness Framework, zero-day, GPT-5.6 https://tpsreport.news/news/world-labs-atlas-3d-world-model TPS — Tokens Per Second en 2026-09-02T12:36:01.356Z World Labs Launches Atlas, an Omni-Model That Generates and Reconstructs 3D Worlds From a Handful of Photos World Labs, Atlas, 3D generation, world models, spatial intelligence, robotics simulation, Fei-Fei Li https://tpsreport.news/news/gemini-agent-based-video-analysis-token-reduction TPS — Tokens Per Second en 2026-09-02T08:35:43.997Z Google Adds Agent-Based Video Analysis to Gemini Flash, Cutting Token Usage by Up to 88 Percent Google, Gemini, video analysis, Gemini Flash, agentic AI, API, YouTube https://tpsreport.news/news/claude-fable-mythos-5-1-cache-price-cut TPS — Tokens Per Second en 2026-09-02T08:06:02.282Z Anthropic Releases Claude Fable 5.1 and Mythos 5.1, Cuts Cache Pricing 75% But Output Tokens Jump 70% Claude, Anthropic, Fable 5.1, Mythos 5.1, pricing, benchmarks, Artificial Analysis, agentic AI https://tpsreport.news/news/z-ai-glm-flash-latest-openrouter-alias TPS — Tokens Per Second en 2026-09-02T01:50:46.516Z OpenRouter Adds Auto-Updating Alias for Zhipu AI's GLM Flash Model Family Z.ai, Zhipu AI, GLM Flash, OpenRouter, multimodal, long context, pricing https://tpsreport.news/news/claude-fable-5-1-reasoning-levels-benchmark TPS — Tokens Per Second en 2026-09-02T00:05:58.020Z Anthropic Releases Claude Fable 5.1, Claims 52.6% on New Terminal-Bench-Science Benchmark Anthropic, Claude, Claude Fable, reasoning models, benchmarks, Terminal-Bench-Science, pricing https://tpsreport.news/news/meta-muse-voice-transcribe-real-time-model TPS — Tokens Per Second en 2026-09-01T22:05:52.470Z Meta Launches Muse Voice Transcribe, Real-Time Speech Model Handling 20+ Speakers Across 70+ Languages Meta, speech-to-text, audio AI, transcription, Meta Superintelligence Lab, multilingual AI, speaker diarization https://tpsreport.news/news/benchmirt-llm-benchmark-auditing-ai2 TPS — Tokens Per Second en 2026-09-01T21:51:04.565Z Ai2 Introduces BenchMIRT, a Method to Reveal What LLM Benchmarks Actually Measure Ai2, LLM benchmarks, evaluation, item response theory, AI safety, research https://tpsreport.news/news/openai-astra-cybersecurity-model-exploitbench TPS — Tokens Per Second en 2026-09-01T21:20:49.126Z OpenAI's Astra Model Aces Cybersecurity Benchmark, Found Two Zero-Day Exploits Unassisted OpenAI, Astra, cybersecurity, AI safety, zero-day, model release, alignment https://tpsreport.news/news/anthropic-claude-watermark-verification-api TPS — Tokens Per Second en 2026-09-01T20:51:06.763Z Anthropic Launches API to Detect Watermarked Text From Claude Models Anthropic, Claude, AI watermarking, EU AI Act, AI detection, compliance, SynthID https://tpsreport.news/news/openai-astra-delay-hugging-face-hack TPS — Tokens Per Second en 2026-09-01T20:50:45.015Z OpenAI Delays Unreleased 'Astra' Model, Says It Cleared First-Ever 'Critical Cybersecurity Capability' Threshold OpenAI, AI Safety, Cybersecurity, Astra, Hugging Face, Model Delay https://tpsreport.news/news/openai-astra-critical-cybersecurity-threshold TPS — Tokens Per Second en 2026-09-01T20:35:46.830Z OpenAI Says Upcoming Astra Model Is First to Cross 'Critical' Cybersecurity Risk Threshold OpenAI, Astra, cybersecurity, Preparedness Framework, AI safety, model release https://tpsreport.news/news/claude-fable-5-1-mythos-5-1-launch TPS — Tokens Per Second en 2026-09-01T20:21:10.925Z Anthropic Releases Claude Fable 5.1 and Mythos 5.1, Cuts Agentic Costs by Up to 45% Anthropic, Claude, Claude Fable 5.1, Claude Mythos 5.1, agentic coding, AI pricing, watermarking https://tpsreport.news/news/anthropic-fable-mythos-5-1-cheaper-less-restrictive TPS — Tokens Per Second en 2026-09-01T19:50:48.901Z Anthropic Releases Fable and Mythos 5.1, Cuts Token Costs and Loosens Safeguard False Positives Anthropic, Fable, Mythos, Zero Data Retention, AI safety, model release