Welcome to Import AI, a newsletter about AI research. Import AI runs on arXiv, cappuccinos, and feedback from readers. If you’d like to support this, please subscribe. Subscribe now LLMs are very bad at low-latency tasks:…Realtime-Bench say…
Firehose
Filtered to tagged “large language models” · clear filters
Browse: People · Companies · Papers · Podcasts · Hacker News · Deep dives
Browse by tag
artificial intelligence 70continual learning 27AI 18open-weight models 15reinforcement learning 14agentic coding 12AI agents 11large language models 9tech 9AI safety 7benchmarks 7finance 7open-source 7robotics 7language models 6machine learning 6Reinforcement learning 6Diffusion models 5game development 5natural language processing 5Agentic AI 4anthropic 4computer vision 4cybersecurity 4data integration 4deep learning 4Recursive self-improvement 4Vercel 4video games 4AI infrastructure 3
Mistral AI has released Le Chonk, a 1-trillion-parameter multimodal model, now available with improved coding, cybersecurity, and reasoning efficiency, leading open-weight benchmarks in coding and defensive security, and offering three reasoning modes for cost-performance balance. AI summary
Mistral Large 4, a multimodal, multilingual, Mixture-of-Experts (MoE) model, takes text and images as input and generates text, with 1T total parameters, 49B active parameters, and a context window of 1M tokens. AI summary
This paper investigates the problem of training-inference mismatch in reinforcement learning for large language models, where the training and inference processes can produce different results, leading to suboptimal training. A practitioner might care about this issue because it can lead to poor performance and instability in training, especially when using off-policy methods.
In general, LLM performance is incredibly multidimensional. There's no guarantee of a fixed ordering where the model that is best at math is also the model that's best at jokes. Despite that complexity, much of the variation in performance …
Katelyn Lesse and Angela Jiang join Latent Space to explain the engineering behind the Claude Platform: launching Opus 5.5, separating the model’s “brain” from the agent’s “hands,” scaling APIs, and building reliable managed agents. A new i…
This paper helps prevent large language models from becoming unsafe by identifying training data that causes them to behave badly, even if the data itself looks harmless. Practitioners can use this technique to build safer language models that are more reliable and trustworthy.
The GPT-6.1 Terra model is a fast and efficient text model for visual reasoning, precise computer use, and clear, natural writing, with a 1,050,000 context window and 128,000 max output tokens. Pricing is based on the number of tokens used, with text tokens costing $1.00 per million tokens and cached input tokens priced at 5% of the uncached input token rate. The model supports various endpoints, including Live, Chat Completions, Responses, Realtime, and Fine-tuning, with a maximum rate limit of 15,000 requests per minute (RPM) for the Grow tier. AI summary
OpenAI and Anthropic have cut their Large Language Model (LLM) prices by up to 80% in a few months, likely due to the commoditization of LLMs and increased competition from open-weight models and secondary inference providers. This trend is driven by the decreasing cost of using LLMs, which is correlated with improvements in model performance and the emergence of more efficient models. AI summary