最大2倍高速な推論エンジン「Magnitude」公開、ハードごとに自動最適化

AI summary

Editorial policy

Magnitude AI released "Magnitude," an open-source LLM inference engine developed for AI agents.

Key points

•It is an open-source LLM inference engine developed for AI agents.
•It is licensed under Apache-2.0, allowing free use with no token costs.
•It features automatic optimization for each hardware configuration.

#LLM / Model Trends #AI Agents#Magnitude

Read the original

Hacker News10/1/2026

Launch HN: Magnitude (YC S25) – Self-optimizing inference engine for agents

エージェント向けにハードウェア上で最大限の速度で動作する自己最適化推論エンジン「Magnitude」が発表され、llama.cppより最大2倍高速であるとされている

PC Watch10/16/2026

長期記憶でOpus 5超えをうたう「Hy-Memory-30B-A3B」、無償公開

Tencent Hunyuan released the open-weight model Hy-Memory-30B-A3B, specialized in creating AI agent memories, which outperforms Claude Opus 5 in memory creation scores.

gihyo.jp10/16/2026

Meta、オープンウェイトモデル「Muse Spark Open」を公開

Meta announced the release of the open-weight model 'Muse Spark Open' on October 15, 2026.

CodeZine10/16/2026

GitHub、複数モデルを組み合わせる「HydraFusion」にGPT-6 Astra、Opus 5.5など追加

GitHub added new models such as GPT-6 Astra and Opus 5.5 to its 'HydraFusion' feature, which combines multiple models.