Mistral Launches Robostral Navigate, 8B Model for Robot Navigation
Mistral's Robostral Navigate, an 8B model, achieves 79.4% success rate on R2R-CE benchmark using just one camera.
ARTICLES BY ALEX LINDGREN
LATEST NEWS
-
-
xAI's Grok 4.5 Costs Less Than Fable 5 and GPT 5.5
xAI's Grok 4.5 scores 83.3% on Terminal Bench 2.1, nearly matching GPT 5.5's 83.4%, but trails Fable 5 by one point.
-
Google updates Android AI dev benchmark with Fable 5 and other models
Google has expanded its Android Bench benchmark with eight new models, including Fable 5, to evaluate AI agents in app development.
-
OpenAI Launches GPT-Live for Real-Time ChatGPT Conversations
OpenAI introduced GPT-Live, a new model allowing ChatGPT to listen and speak simultaneously, improving conversation flow with 75.7% user preference over previous versions.
-
SpaceXAI Launches Grok 4.5, Its Smartest Model for Coding and Knowledge Work
SpaceXAI launched Grok 4.5 on July 8, 2026, claiming it outperforms leading models on multiple benchmarks, including scoring #1 on Harvey's Legal Agent Benchmark.
-
OpenAI Launches GPT-Live for Natural Voice Interaction
OpenAI introduced GPT-Live, a new voice model, on July 8, 2026, enhancing ChatGPT Voice with full-duplex capabilities and improved conversational flow.
-
Mistral Unveils Robostral Navigate for Embodied Navigation
Mistral's Robostral Navigate, an 8B model, achieves 76.6% success on unseen R2R-CE benchmarks using only a single RGB camera.
-
Cohere Launches Open-Source Arabic Speech-to-Text Model
Cohere released Cohere Transcribe Arabic, the most accurate open-source Arabic speech-to-text model, achieving a 25.87 WER on the Hugging Face Arabic ASR Leaderboard.
-
Meta Launches Muse Image Model With Privacy Concerns
Meta's Muse Image model ranks second in image generation benchmarks, but its use of Instagram photos without consent has sparked privacy debates.
-
Chinese AI Startup MiniMax to Open-Source 2.7 Trillion Parameter Model
Chinese AI startup MiniMax plans to open-source a 2.7 trillion parameter model by late 2026, surpassing current Chinese models in scale.
-
NVIDIA Releases GR00T 1.7 Vision-Language-Action Model on LeRobot
NVIDIA's GR00T 1.7, a Vision-Language-Action model, is now available on LeRobot, enabling developers to train and deploy humanoid robots more efficiently.
-
OpenAI Unveils GPT-5.6 Models After U.S. Government Delay
OpenAI's GPT-5.6 models launch Thursday after a U.S. government-mandated delay. The Department of Commerce approved the public release following additional testing.