GoKawiil
Tech News
OpenAI has released GPT-6 Astra, a new flagship model arriving less than two months after its GPT-5.6 family, which the company calls its most intelligent and aligned system yet. Astra is built to handle multi-step computer-use tasks, coding, browsing and professional workflows simultaneously, and OpenAI backs the claims with strong scores on benchmarks like ARC-AGI-3, Terminal Bench 4.0, and the Agent's Last Exam.
OpenAI's unreleased GPT-6 Astra model scored 62.7% on the ARC-AGI-3 Semi-Private benchmark using a Standard harness for $26,000, and 99.9% using a Provider Adapter harness that preserves reasoning state between requests for $19,000. The model also beat the median human tester's action efficiency on 96% of the benchmark's levels, building compact symbolic world models and its own shorthand notation to track game states.
Russian startup Mostik has developed a method allowing AI models to exchange information through their internal weight values rather than generated text, effectively letting a smaller model absorb capabilities from a larger one. The team demonstrated this by linking a 753-billion-parameter GLM-5.2 model with a 4-billion-parameter Qwen-3.5 model, producing a hybrid system that runs at one-twentieth the cost of the full-size model while performing roughly midway between the two in capability. Mostik also used a related technique to build a model that has topped the ARC-AGI 3 benchmark, though details remain undisclosed while the contest is ongoing.
A researcher built and trained a small transformer model from scratch in 1.5 hours on a single RTX 5090 GPU, achieving 44% accuracy on the ARC-AGI-1 public evaluation set and 7% on ARC-AGI-2. The model, an upgraded and cheaper version of a previous open-source release, matches the performance of existing test-time-training approaches like TRM and HRM while using far less compute.