Last 90 days · 2821 total
Grok Build is now in Beta for SuperGrok and X Premium+ users, offering features like Plan Mode, image/video generation with Imagine, and automation b…
OriginalMistral Vibe has officially launched as an AI agent specializing in long-horizon productivity and coding, featuring Work mode, Code mode, a CLI, and…
OriginalThe llama.cpp project has successfully merged support for Gemma-4 MTP, indicating enhanced integration and performance for the model within the frame…
OriginalResearchers introduce Lean4Agent, a method for formal modeling and verification aimed at enabling Large Language Models (LLMs) to execute reliable mu…
OriginalA user's benchmark suggests that the Gemma4_31b_fp8 model can perform comparably to Sonnet_4.6_medium in tasks such as Cypher queries and entity extr…
OriginalNew research advocates for re-centering human involvement in evaluating LLM personalization, moving beyond the current reliance on synthetic data.
OriginalThis research characterizes distinct token-level signatures left in reasoning traces, offering insights into how language models commit and persist i…
OriginalBrowse.sh has been launched on Product Hunt, offering a solution to provide AI agents with "muscle memory" for automating web tasks.
OriginalGOPAgen is a new framework designed for motion-aware and efficient agentic long-video understanding, leveraging structural memory and hierarchical re…
OriginalElmes* introduces an automated method for constructing fine-grained evaluation rubrics, specifically for Large Language Models operating in long-tail…
Original