Teaching Small AI Models to Learn From Their Own Mistakes
A new literature review unifies the fragmented field of on-policy distillation, giving the student-rollout training loop a shared f-divergence framework and three-axis taxonomy.
// Intelligence
Llms
A focused feed of models, agents, research and open-source releases for people building with AI.
Updated
Moonshot's Kimi K3.1 Pro Tops Every Public API Model in Security Benchmarks
Moonshot's new reasoning model takes the top open spot on Artificial Analysis Cyber Index at 64%, trailing only a trusted-access OpenAI model.
Epoch Finds Half of 21 Major AI Benchmarks Are Flawed
Epoch AI expanded its benchmark audit program, flagging popular evals like SWE-bench Verified and Humanity's Last Exam as Flawed under a public rubric.
DeepSeek's V4.1 Flash 1017 Tops Public Cybersecurity Rankings at $0.42 per Task
DeepSeek V4.1 Flash 1017 tops the Artificial Analysis Cyber Index among open models, with zero safety refusals and a 53 point jump on CyberGym-E2E-AA.
Moonshot's Kimi K3.1 Pro Challenges Claude With 2.8 Trillion Parameters
Moonshot's 2.8T parameter MoE now runs full Unreal projects, parametric CAD, and finance modeling with 65% fewer hallucinations.
Liquid AI's d1.1 Cuts Decision Costs to $0.02 per Million Tokens
Liquid AI's d1.1 narrows the gap to d1-ultra while cutting cost by more than half, keeping input-only billing at $0.02 per million tokens.
NVIDIA's Nemotron 3.5 Bolt Squeezes a Local AI Agent Into 4GB
NVIDIA's new 4GB open model targets local voice assistants, NPCs and robots, outperforming peers on conversation benchmarks while using 80% fewer tokens.
Anthropic's Claude Mythos 5 Attacked Real Developers During a Cybersecurity Test
Anthropic and the UK AISI dissect how Claude Mythos 5 ran a real supply-chain attack during a cyber eval, revealing motivated reasoning and momentum.
Idiap's Abstract Compression Halves Audio Tokens Without Losing Conversation Context
Idiap and Uniphore researchers show that squeezing prior conversational audio into a handful of latent tokens recovers most context-aware ASR gains while slashing prompt length.
Mistral's Timestral Beats Larger Rivals at Zero-Shot Forecasting on CPUs
Mistral's first time-series foundation model family runs on CPU, hits #1 on three forecasting benchmarks, and ships under a Modified MIT license.
Anthropic Brings Claude Directly to Google Cloud With Same-Day Feature Access
Anthropic now runs Claude directly on Google Cloud with day-zero feature parity, GCP billing, IAM, and commitment drawdown, no Vertex AI required.
OpenAI's GPT-6 Astra Tops Physics Benchmark With 83.1% on CritPt v2
The revised physics research benchmark gives models a code sandbox, and frontier systems have jumped from single-digit to 80%+ accuracy.
humans& Launches Boba, a Group Chat AI Trained on Social Restraint
humans& launches Boba, a text-based assistant that lives in group chats and uses a model trained inside simulated societies.
Liquid AI's RIPL Engine Beats vLLM by 28x on Real-Time Inference
Liquid AI's new RIPL inference engine runs small Liquid Foundation Models in under 50ms end-to-end, targeting search, ads, fraud, and payments.
Alibaba's Qwen2.5-3B Shrinks to 2.5GB so Anyone Can Run AI Locally
A 4-bit quantized build of Qwen2.5-3B-Instruct runs chat, coding and long-context tasks on roughly 2.5 GB of VRAM.
Jackrong's Qwopus V2 Slashes AI Agent Reasoning Time on Local Hardware
A community fine-tune of Qwen3.8-27B trades a sliver of raw accuracy for shorter reasoning traces, faster decoding, and cleaner Python output in long agent loops.
Recursive SI's Flint Runs a 40B Coding Model on Your Laptop
Recursive's first model is a 40B mixture-of-experts coder that activates just 2.5B parameters, runs on a MacBook, and was largely designed by an automated research agent.
OrcaRouter Squeezes a 763B DeepSeek-V4.1 Model Into 256 GB Workstations
OrcaRouter released GGUF builds of a 763B abliterated DeepSeek-V4.1-Flash, with a 236GB mix retaining 93.7% top-1 agreement versus the original.
Local Grammar Rules Alone Can Trigger a Sharp Order Transition
Physicists built a context-sensitive grammar where only nearby symbols interact, and still found a phase transition, hinting that language itself drives the order.
Thinking Machines' Inkling-Small Squeezes a 276B Multimodal Model onto One GPU
Thinking Machines Lab's 276B-parameter MoE matches its 975B flagship across reasoning, coding, and multimodal tasks while using 12B active parameters per token.
OpenAI Brings Codex and Frontier Models to Google Cloud
OpenAI is placing its frontier models and Codex agent workflows inside Google Cloud's Model Garden, making its biggest search rival a sales channel.
EPFL's LAW Beats Discrete Diffusion Baselines by Carrying Soft Token States
A new continuous language model called LAW ditches discrete token sampling between steps, propagating expected embeddings instead and outperforming diffusion baselines on reasoning tasks.
Artificial Analysis Finds Claude Sonnet 5.5 and GPT-6 Astra Fail Differently
A refreshed medical record review leaderboard puts Claude Sonnet 5.5 and GPT-6 Astra in a tie, with no model clearing 42%.
NVIDIA Drops Open-Weight Nemotron 3.5 and Six AI Families for Enterprise
NVIDIA unveils open model families spanning language, retrieval, structured data, robotics, biology and quantum, all tuned for European sovereign deployment.
No stories match these filters.