All News

1533 technical news articles

IBM Granite 4.2 Opens Three Dense Reasoning Models, but Official Documents Disagree on Their Training Starting Point

Granite 4.2 introduces 3B, 8B, and 30B models with switchable thinking modes, native tool calling, and context windows of up to 512K, under the Apache 2.0 license. The 8B and 30B models underwent sandbox-based agentic reinforcement learning, but the model card and technical article contradict each other on whether the models were pretrained from scratch.

2026-08-26

代理工具

Pipette Open-Sources On-Device Model Benchmarking for Accuracy, Latency, and Memory—but Rankings Still Cannot Be Compared Directly Across Devices

Liquid AI and Artificial Analysis have released Pipette, which measures on-device inference across more than 1,000 combinations of models, quantizations, runtimes, devices, and context configurations. Speed and memory are measured on physical devices, while quality scores are generated separately on H100 GPUs, so the two testing paths must be distinguished when interpreting its combined charts.

2026-08-26

AI 安全

TRACE Joins the Linux Foundation: Using Hardware Attestation to Prove Where an Agent Runs, but Not That Its Behavior Is Correct

The Linux Foundation has accepted the TRACE specification, which creates portable cryptographic evidence for AI agents’ models, execution environments, policies, and tool calls. The current v0.2 remains a Developer Preview: it can prove that records are linked to an attested environment, but not that an agent completed its task or that the policy itself is secure.

2026-08-26

代理工具

TRACE Rewrites Agent Skills from Successful and Failed Trajectories, Raising the Consistent Pass Rate on Hidden Tests to 70%

Xiaomi’s team introduced TRACE, which iteratively updates a retrievable Skill Bank by contrasting successful and failed trajectories without modifying the underlying model weights. It increased the share of CAR-bench hidden-set tasks passed in all three runs, but also raised average token usage, cost, and latency.

2026-08-25

代理工具

Gradio Turns AI Pipelines into Executable Node Graphs, but Canvas and API Concurrency Semantics Still Differ

Gradio has officially introduced the built-in `gr.Workflow`, allowing the same node graph to be used for visual debugging, deployed to Spaces, and automatically exposed through REST endpoints. The interactive canvas can execute nodes at the same level in parallel, but generated API calls currently still run sequentially, so canvas performance results cannot be applied directly to API testing.

2026-08-25

代理工具

Dify 1.17 Packages Agent Environments as Restorable Snapshots While Overhauling the Runtime Backend and Environment Variables

Dify 1.17.0 adds the E2B runtime backend, build-time home snapshots, versioned Skills, and layered context compression, giving agents a consistent set of tools and file state on every startup. This is not a painless minor upgrade: database migrations, Compose configuration, and multiple environment variables have changed, so self-hosted users must revalidate their deployments.

2026-08-25

GitHub Repo

Headlong Implements a Persistent Agent With an Infinite Bash Loop, Thinking and Acting Even When No One Is Asking

Laude Institute and MIT have open-sourced Headlong, a Bash-based system with fewer than 10,000 lines of code that treats messages as asynchronous observations within a continuous stream of thought rather than as new sessions. The design enables research into long-term memory and autonomous action, but it also turns token costs, permission isolation, and cross-user data boundaries into persistent risks.

2026-08-25

應用研究／科學資料工程

Astro-COLIBRI Parses Astronomy Circulars with a Constrained LLM, Turning a Decade of Free Text into 68,393 Observations

A new pipeline converts photometry, redshift, and observation times from NASA GCN Circulars into event-level structured data and is already operating in real time. An internal audit recorded 99.80% field-level accuracy, but time interpretation and observatory attribution remain the main sources of error, so researchers must consult the original text.

2026-08-25

NVIDIA Adds Two-Stage Generation for MiniMax H3, Cutting 10-Second Video Latency to 14.9 Seconds on a Single GB200

H3 Super Acceleration first creates a low-resolution draft with four denoising steps, then uses LTX and Sol-Attn to complete high-resolution refinement in three steps. NVIDIA reports that 768p, 10-second video generation is 27.7× faster than the SGLang baseline, but the acceleration changes the sampling path and is not a pure kernel optimization at equivalent quality.

2026-08-25

代理工具

Agent Lightning 1.0.1 Packages the Agent-Tuning Workflow as a Skill, Requiring Per-Run Measurement of Cost, Latency, and Accuracy

Microsoft has introduced an Agent Lightning Skill for Claude Code, Codex, and GitHub Copilot that lets a coding agent modify another editable agent against a fixed benchmark. It provides a budget-constrained set of experimentation rules; it does not automatically perform reinforcement learning on the target agent.

2026-08-25

推論系統

Ray 2.58 Completes KV-Cache/Token-Aware Routing, but the New Router Remains Alpha

Ray Serve LLM now centralizes prompt tokenization, cache-hit estimation, and replica selection at the ingress, then passes tokens directly to the inference engine. The design can reduce repeated prefill and tokenization, but no updated performance data has been published, and the control plane introduces additional synchronization overhead.

2026-08-25

推論系統

AWS Uses a Small Model to Compress RAG Evidence First: Input Tokens Cut to One-Tenth, but Request Latency Rises 12%–19%

The new architecture adds a low-cost model call between retrieval and the primary model, retaining only verbatim passages relevant to the question and their source identifiers. AWS’s internal tests recorded a roughly 30% cost reduction, but the evaluation relies on a single enterprise corpus and an LLM judge.

2026-08-25