Terminal Agents Stumble, Reward Hacking, and a Fix
New arXiv research shows agents pass just 15.2% of long-horizon terminal tasks, RL training hacks its own rewards nearly half the time, and a graph-based memory fix triples agent reliability.
Articles Tagged "Benchmarks"
Mistral's Leanstral 1.5 Finds 5 Unknown Bugs Free
Mistral's open-source Leanstral 1.5 scanned 57 repos and found five previously unreported bugs, including a silent integer overflow in a Rust zigzag decoder.
Grok 4.5 Review: Agentic Speed at Half the Price
xAI's Grok 4.5 tops AutomationBench and costs 80% less per agentic task than Opus 4.8, but neutral coding benchmarks and a doubled hallucination rate complicate the story.
ByteDance Seedream 5.0 Pro - 2x Faster Than GPT-Image 2
ByteDance's Seedream 5.0 Pro reaches developer APIs with 2K output, 14-language text rendering, and layer-based editing at roughly one-fifth the cost of GPT-Image 2.
AI Research: Smarter Evals, Token Cuts, Agent Gates
Three papers tackle benchmark saturation, orchestration waste, and silent policy violations in tool-using agents.
Grok 4.5
Grok 4.5 is xAI's 1.5-trillion-parameter V9 MoE model, publicly launched July 8 at $2/M input - cheap, fast, and token-efficient, though neutral harness benchmarks put it well behind Fable 5 and Opus 4.8 on coding.
Grok 4.5 Ships - Token Efficiency Is the Real Edge
Grok 4.5 goes public with real benchmarks: behind Fable 5 on coding evals, but a 4.2x token efficiency gap that changes the cost math for high-volume pipelines.
Best AI for Creative Writing - July 2026
Claude Fable 5 tops EQ-Bench Longform at Elo 2189 while GPT-5.5 leads the Mazur Writing Benchmark, reshaping the creative writing model rankings in July 2026.
GPT-Live-1
OpenAI's full-duplex voice model that listens and speaks simultaneously, replacing Advanced Voice Mode in ChatGPT with three reasoning tiers backed by GPT-5.5.
How AI Agents Break - Plus Fixes for Memory and Tools
Three arXiv papers map how LLM agents fail across 19 benchmarks, show in-process memory cuts retrieval latency 1,000x, and reveal steering vectors that control tool invocation.
Research Replication, Safe Responses, Verifiable Reasoning
Three new papers tackle AI verification from different angles: automated scientific replication, constructive safety alignment, and neurosymbolic reasoning programs.
ZCode Launches Free - GLM-5.2 Comes for Claude Code
Z.ai's free agentic IDE ships Goal Mode, multi-agent coordination, and pricing up to 82% cheaper than Claude Code - with a concrete China data law risk that teams need to weigh.
