뉴스 및 전체 검색
뉴스, 도구, 모델, 가이드, 주제, 워크플로, 프롬프트를 함께 검색합니다.
뉴스 기사 98
“AI Code Guide”
AI 에이전트는 미디어의 새로운 최애 타깃이 되다
에이전트가 대중화됨에 따라, 일부 출판사들은 사람들이 인간이 아닌 AI를 위해 맞춤 제작된 콘텐츠에 비용을 지불하기를 기대하고 있습니다.
b11922
tests : fix multi-seq split replay and partial-state cases for trained models (#31216) * tests : fix multi-seq split replay against trained models The rollback snapshot is captured by the graph that decoded the rolled-back tail, while the reference context decodes the same prefix with different batch shapes. The two states differ in the last bits and trained models amplify the difference with every replayed token. Compare only the first replayed token of each seq, then restore the rollback state into the reference context before the seq-1-only tail. Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB) * tests : require bit-exact seq-1 isolation in split replay The seq state save is not always bit-lossless - SWA-masked cells are not saved - so reload the rest
b11928
model: Add MiniCPM-OCR-Jev (#30980) * model : add MiniCPM-OCR-Jev support The model is a Qwen3.5-VL trunk with a value head read at the last token: it scores an OCR text against an image rendered from it, it does not generate. Add the openbmb-minicpm-ocr-jev decision type, served by POST /v1/systemone. The class scores are the pooled output of the sequence, read with last pooling, so the conversion turns score.weight into cls.output and sets embedding_length_out. The checkpoint has no LM head, so text generation is rejected instead of using the output weights duplicated from the token embeddings. The task is fixed, so its prompt is entirely in the systemone template, the same way h2o-lightning keeps its default there. The processor of the checkpoint allows up to 16384 merged tokens per ima
듀얼페이스: LLM 서비스를 위한 분리형 스케줄링 및 협력적 GPU–CPU 디코딩
arXiv:2610.33501v1 발표 유형: 신규 초록: 대규모 언어 모델(LLM) 서비스는 추론 처리량을 높이기 위해 배치 처리에 의존하지만, 키-값(KV) 캐시는 시퀀스 길이와 함께 증가하여 GPU 상주 동시성을 제한한다. CPU 오프로드를 통해 호스트 메모리를 활용하면 KV 용량을 확장할 수 있지만, 기존 시스템은 여전히 대부분 GPU 중심으로 작동한다: CPU 상주 요청은 GPU 스케줄러가 적절한 겹침 기회를 제공할 때에만 진행된다. 그 결과, 호스트 메모리와 CPU 연산은 독립적인 서비스 자원으로 작용하지 못하며, 이로 인해 부하 하에서 CPU 측 진행이 간헐적으로 발생하고 꼬리 부분의 토큰 간 지연(ITL)이 부풀어 오르게 된다. 우리는 호스트를 독립적으로 스케줄링되는 디코딩 엔진으로 전환하는 LLM 서비스 런타임인 DualPace를 제안한다. DualPace는 두 개의 엔진 간에 스케줄링 권한을 분리한다: GPU 엔진
Coding From the Himalayas With an Offline AI Agent
How to set up and run an offline coding agent using OpenCode, Ollama and Qwen.
RSI-Opt: Recursive Self-Improvement Optimizer Agents For Model Training
arXiv:2610.32026v1 Announce Type: new Abstract: Conventional neural-network training typically follows a predefined optimization recipe, motivating a LLM agent that monitors training dynamics and adjusts optimizer settings as training progresses, but a fixed agent policy does not accumulate experience across runs. We introduce RSI-Opt, a plug-and-play Recursive Self-Improvement Optimizer Agents which evolves the knowledge that guides this process: a prompt, a playbook of procedures, a library of tested optimizer code, and a decision cadence. A meta-agent rewrites this policy using evidence from earlier runs, and an archive retains evaluated policies for further evolution. Across 5 generations on three training tasks, the meta-agents develop reusable procedures and code and conduct their ow
b11889
cuda : use std::string instead of VLA for archName (#30463) * cuda : use std::string instead of VLA for archName Modernize the code (old C style) and fix the following warnings: ``` ggml/src/ggml-cuda/ggml-cuda.cu:180:19: warning: variable length arrays in C++ are a Clang extension [-Wvla-cxx-extension] 180 | char archName[archLen + 1]; | ^~~~~~~~~~~ ggml/src/ggml-cuda/ggml-cuda.cu:180:19: note: read of non-const variable 'archLen' is not allowed in a constant expression ggml/src/ggml-cuda/ggml-cuda.cu:179:9: note: declared here 179 | int archLen = strlen(devName); | ^ ``` Signed-off-by: Adrien Gallouët * Add a comment about archName Signed-off-by: Adrien Gallouët --------- Signed-off-by: Adrien Gallouët Signed-off-by: Adrien Gallouët **Website:** - **Attestations:** - **macOS/iOS:** - [ma
b11895
metal : stop overflow from happening in large matrix output offsets (#30072) * Metal : stop overflow from happening in large matrix output offsets Assisted-by: Codex * metal : widen batch offsets and MUL_MAT_ID token indices Assisted-by: Codex --------- Co-authored-by: feni6 <270037094+feni6@users.noreply.github.com> **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11895/llama-b11895-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b11895/llama-b11895-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/dow
Context Engineering (Agent Memory) for AI Agents Starts with Retrieval
Agents waste time and context on clumsy searches, oversized results and stale data. Learn the retrieval techniques LangChain's Deep Agents harness uses to get the right context in.
b11888
common, grammar: add recursive depth limit, make it a requirement (#31153) * common, grammar: add recursive depth limit, make it a requirement * catch more recursive functions * common : tidy up peg parser changes * grammar : reject candidates in a loop, without a depth limit The depth limit in llama_grammar_reject_candidates_for_stack counted the code points of a vocab token, so any token of 256 chars or more was always rejected under a grammar, such as the 512-space token of GLM-4.7-Flash or the long space tokens of Laguna-S-2.1. The rejection now walks the code points with its own stack of levels instead of recursing: there is no limit and no stack risk, whatever the length of a token from the GGUF, and the masks match master. * grammar : track rejected candidates as indices into the in