A self-editing search agent with 20B parameters published by Chroma. It performs multi-hop search while dynamically pruning the context, and shows the same or higher accuracy than the Frontier model at 1/10 the cost and up to 10 times faster latency. Weights are exposed in Apache 2.0.
Hypura breaks away from llama.cpp’s mmap design and streams even dense models with a three-tier NVMe placement, while TurboQuant eliminates quantization-constant overhead via a polar-coordinate transform. Includes a design comparison with Flash‑MoE and a review of scenarios where KV‑cache compression actually helps.
LiteLLM 1.82.7 and 1.82.8 were poisoned on PyPI for about 46 minutes. TeamPCP stole a PyPI token through Trivy's CI/CD and injected malware that collects more than 50 credential types, including SSH keys, AWS, Kubernetes, and Docker secrets.
The Cosmos 2.5 series world model announced by NVIDIA at GTC 2026 is mainly for industrial use, but it has reached the stage where the 2B parameter model can be run on the Jetson Orin Nano, which costs less than $500. We have organized the edge deployment of physical AI, from industrial robots to pet robots.
Composio publishes security analysis of OpenClaw. Approximately 7.1% of SkillHub-distributed skills were found to have critical vulnerabilities, leaving over 30,000 instances exposed to the internet in the early stages at risk of prompt injection and credential theft.
Flash-MoE is a C/Metal inference engine that runs Qwen3.5-397B-A17B on a MacBook Pro M3 Max at 4.36 tokens/s. With expert streaming from SSD and hand-written Metal shaders, it fits the 209GB model into a 48GB memory budget.
The three-stage pipeline of BERT perplexity scan → LLM judgment → escalation packaged as a cross-platform Python tool. The installer automatically downloads llama-server and GGUF models.
Redesigned with inference latency as the first priority, Mamba‑3 combines exponential trapezoid discretization, complex‑valued states, and a MIMO structure to reach about 6.9× the speed of a Transformer at 16,384 tokens.
Compresr's YC-backed Context Gateway is a proxy between AI agents and LLM APIs. Its three pillars - preemptive summarization, tool output compression, and tool discovery - reduce wasted context-window usage.
Sakura Internet's "Sakura AI Engine" is an LLM inference platform compatible with OpenAI API. There is a free limit of 3,000 requests per month, and multiple models such as Kimi-K2.5 and gpt-oss-120b can be used domestically.
Cursor released Composer 2 without disclosing its base model; calling its OpenAI-compatible API revealed it is Kimi K2.5. This escalated into a licensing dispute, but a formal commercial agreement with Moonshot AI was subsequently confirmed.
AttnRes to replace Transformer's fixed residual combination with softmax attention in the depth direction. Demonstration with Kimi Linear 48B improved GPQA-Diamond +7.5pt and HumanEval +3.1pt. Training overhead was kept below 4% and inference below 2%.
