GitHub Repo

GitHub 熱門專案、開源工具、框架、runtime 與重要 release。

ScienceIDE Packages Scientific Software into Verifiable Agent Environments and Feeds Numerical Simulation Results Directly into Training

ScienceIDE combines fixed versions of scientific software, containers, physics cases, and private verifiers into shared environments for SFT, RL, and evaluation. Initial results suggest that scientific repair trajectories can improve some models and benchmarks, but only a small subset of the environments and tasks has been released, leaving limited evidence of generalization.

2026-09-17

推論系統

Chord Rewrites the INT4 MoE Kernel for Kimi K2.x, Selecting GPU Scheduling by Inference Phase

Novita AI and vLLM have open-sourced Chord, a W4A16 CUDA operator tuned for Kimi K2.x serving workloads on H200, B200, and B300 GPUs. Public benchmarks show performance of up to 2.15× that of Humming, but the figures cover only the MoE layer kernel and cannot be directly translated into end-to-end throughput.

2026-09-17

硬體

OpenVINO 2026.4 Brings Multi-Token Draft Decoding to CPUs and GPUs, Unifies NPU Performance Tracing

OpenVINO 2026.4 adds Multi-Token Prediction speculative decoding for Gemma 4, Qwen3.5, and Qwen3.6, while enabling VTune to analyze inference across CPUs, GPUs, and NPUs. The release expands local model coverage, but several NPU dynamic-shape and model-unloading features remain in preview, and Intel has not provided standardized public acceleration figures.

2026-09-17

代理工具

ProgramDistill Evaluates Agents Rebuilding Software from Replayable Web Behaviors, with Full-Workflow Success Rates of Only About 50%

A Microsoft Research team had coding agents observe interactive reference websites and then recreate their functionality without access to the source code. ProgramDistill generated 4,063 tasks from 26 applications, showing that repairing individual features is far easier, while complete workflows dependent on multiple stages of state remain the primary bottleneck.

2026-09-17

代理工具

Stellar Colosseum Uses Dependency Graphs and Falsification Loops to Manage Long-Horizon Mathematical Agents

A team from Google Research and CMU separates strategy exploration, proof decomposition, local repair, and global verification into a multi-agent workflow. Cross-model selection reaches 71% on TCS-Bench, but the main harness has not been fully released, while automated grading and large inference budgets limit comparability.

2026-09-17

AI 基礎設施

NVLink 6 Unifies Link Retry, Fault Isolation, and Hot Standby for Inference in a Single Recovery Stack

NVIDIA has disclosed NVLink 6’s multilayer fault-tolerance design, extending from physical-layer bit-error correction to the NMX Controller and Dynamo inference processes. Internal B200 testing reduced the time required to restore service on two workers after a failure from 283 seconds to 7.3 seconds, though checkpointing tools and NCCL edge cases are still evolving.

2026-09-17

代理工具

GitHub Uses Agents to Rewrite the Copilot Runtime as 830,000 Lines of Rust, Eliminating Per-Client Node.js Overhead

Copilot’s shared agent runtime has migrated from TypeScript/Node.js to native Rust and can be embedded directly into SDKs for six languages through a C ABI. GitHub’s stress tests showed session throughput rising to 120 per second, though the figures come from internal tests that excluded model and network latency.

2026-09-17

LimiX-2 Uses Joint Mechanism Modeling for Tabular Data, With a Single 400M-Parameter Model Spanning Three Task Types

Rather than estimating only conditional labels, LimiX-2 learns the joint structure of features and targets from synthetic causal data. The team has released the weights and inference code, but the model is restricted to non-commercial use, and its leading results come primarily from tabular benchmarks run by the authors.

2026-09-17

代理工具

Foundry Toolkit 1.6.13 Adds Support for Non-ACR Private Container Registries and Changes the Windows ML Runtime

The new VS Code toolkit can deploy hosted agents from private container registries configured with a connection, while also adding an AMD NPU build of Fara 7B and a quantized Aion adapter. Windows ML has been upgraded to `windowsml` 2.3, requiring existing projects to review their execution providers and dependencies.

2026-09-17

推論系統

JustFit Compresses and Dynamically Schedules KV State, Enabling a 24 GiB MacBook to Complete a 210K-Token Workload

By combining 4-bit KV, component residency management, and cross-request state transitions in MLX, JustFit enables a 27B model to perform long-context inference within 24 GiB of unified memory. The reported 6.93× capacity increase comes from a single-machine, model-specific test, and some historical environment details and checkpoint identities cannot yet be fully reproduced.

2026-09-16

代理工具

ScienceBuddy Alternately Rewrites the Agent Harness and Model, Turning Research Interactions into Ongoing Training Material

ScienceBuddy converts researchers’ requests, feedback, and execution logs into tasks and scoring rules, then alternately optimizes the agent code and model. The team has released a simplified experimental framework, but the full code for the hosted workspace and evidence of long-term self-improvement remain unpublished.

2026-09-16

ModAR 先預測軌跡、語意與深度再決定動作，降低世界—行動模型對 RGB 的依賴

CMU 的 ModAR 以共享擴散 Transformer 依序去噪多種未來表徵，再生成機器人動作。小規模實驗顯示，結構化表徵比額外預測 RGB 更有用，但程式碼尚未公開，真實世界證據也只有三項雙臂任務。

2026-09-16