Mistral AI Releases Robostral Navigate: An 8B Model Enabling Robots to Navigate Complex Environments Using a Single RGB Camera
Mistral AI has released Robostral Navigate, its first model built for embodied navigation. The 8B model takes RGB images and a plain-language instruction, then moves a robot. Notably, it reaches 76.6% success on R2R-CE validation unseen using only a single RGB camera. What is Robostral Navigate? Robostral Navigate is an 8B model for robotic navigation through complex environments. These environments include offices, residential buildings, commercial buildings, and outdoor settings. You gi
Get smart on it
An 8B navigation model enables robots to move through complex indoor and outdoor environments by processing a single RGB camera feed and following plain-language instructions. The model uses a "pointing" method to predict where to move next by identifying target locations in the camera's current view, then falls back to local displacement commands when targets are out of view. Training efficiency was achieved through prefix-caching, which reduced training tokens by 22 times and compressed episodes into single forward passes, and the model was further improved through online reinforcement learning that added 3.2% to its success rate. On a standard benchmark for instruction-following navigation, the model achieved 76.6% success on unseen environments using only a camera, outperforming systems that use depth sensors or multiple cameras.
Related news
PrismML Releases Bonsai 27B: 1-bit and Ternary Builds of Qwen3.6-27B That Run on Laptops and Phones
PrismML just released Bonsai 27B. It is a low-bit representation of Qwen3.6-27B, not a new pretrain. The architecture is unchanged. Two variants ship under Apache 2.0. Ternary Bonsai 27B uses {−1, 0, +1} weights at a true 1.71 bits per weight. Its ideal size is 5.9GB. 1-bit Bonsai 27B uses binary {−1, +1} weights at 1.125 bits per weight, for 3.9GB. Both are multimodal. The split is ~24.8B language weights, a 0.46B vision tower, and 2.5B in embeddings and the LM head. The vision tower is
Simulating everything, sort of: The promise and limits of world models
Over the past few years, many of us have gotten a crash course in what we now call artificial intelligence—but really, it has mostly been a crash course in large language models. Increasingly, however, LLMs are no longer the only category of AI drawing high expectations, massive funding rounds, and significant research and product development. Over the past year, we've seen a plethora of new announcements in a category labeled "world models," and you'll likely see more movement there in the comi
