Qwen3.6-Plus
Alibaba's 1M-token flagship agentic coding model posts 78.8% on SWE-bench Verified and undercuts Kimi K2.6 and Claude Opus on price, but ships with no weights and a mandatory reasoning tax.
Articles Tagged "LLM"
Hermes 4.3
Nous Research's 36B open-weight model matches Hermes 4 70B on most benchmarks, tops RefusalBench on alignment, and is the first production model trained entirely on the Solana-secured Psyche network.
CoT Attacks, Weight Secrets, and the Finetuning Gap
Three new papers expose how AI safety monitors can be manipulated, how reasoning weights leak training secrets, and why fine-tuned models fail to use what they know.
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.
OpenAI Launches GPT-Live, Bets Voice Beats Text
OpenAI launched GPT-Live-1 and GPT-Live-1 mini on July 8, replacing Advanced Voice Mode and making an explicit bet that voice will become AI's primary interface.
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.
JadePuffer Shows How AI Agents Now Run Ransomware
Sysdig documents the first AI-agent ransomware operation: an LLM exploited CVE-2025-3248 in Langflow, moved laterally, and encrypted 1,342 production database records with no human directing each step.
Grok 4.1 Fast
Grok 4.1 Fast is xAI's agent-optimized model with a 2M-token context window, #1 ranking on tau-bench Telecom, and one of the lowest input prices among frontier-adjacent APIs at $0.20/M tokens.
GPT-5.4 mini
OpenAI's mid-range model in the GPT-5.4 family delivers near-flagship coding and agentic performance at $0.75/M input tokens with a 400K context window.
Agent Phase Collapse, Reasoning Exits, Preference Gaps
Three new arXiv papers map capability cliffs in agent world models, the narrow benefit of learned reasoning stops, and a 56% accuracy ceiling when agents help users build preferences.
Agent Languages, Sampling Ceilings, and Abstention
Three new papers on agents inventing symbolic languages to cut reasoning tokens by 3-6x, sampling ceilings that waste inference compute, and context-engineering to double agentic abstention rates.
