我的收藏提交工具
FOXAI SEARCH

全站搜索

新闻报道 231

Release: v5.21.0

# Release v5.21.0 ## New Model additions ### MiniCPM-OCR MiniCPM-OCR is a document-understanding model from OpenBMB's MiniCPM-V family. It keeps the MiniCPM-V 4.6 architecture, which pairs a SigLIP vision encoder and a window-attention merger with a Qwen3.5 language model. It is specialized for reading text-dense images such as scanned pages, screenshots, receipts and tables. **Links:** [Documentation](https://huggingface.co/docs/transformers/main/en/model_doc/minicpm_ocr) | [Paper](https://huggingface.co/papers/2509.18154) * Feat test (#49920) by @guarin in [#49920](https://github.com/huggingface/transformers/pull/49920) ### Flint Flint is Recursive's mixture-of-experts language model. Compared to a standard SwiGLU MoE transformer such as Qwen3-MoE, it uses hybrid attention (sliding-windo

翻译服务暂未运行,当前显示原文。
变压器发布

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

翻译服务暂未运行,当前显示原文。
llama.cpp 发布

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

翻译服务暂未运行,当前显示原文。
llama.cpp 发布

DualPace:用于大语言模型推理的解耦调度与GPU-CPU协同解码

arXiv:2610.33501v1 发布类型:新 提要:大型语言模型(LLM)的推理服务依赖于批处理以提升推理吞吐量,但键值(KV)缓存会随序列长度增长,从而限制了GPU上的并发度。通过CPU卸载,可以利用主机内存扩展KV缓存容量,然而现有系统在很大程度上仍受GPU瓶颈制约:只有当GPU调度器能够提供合适的重叠机会时,CPU驻留的请求才能取得进展。因此,主机内存与CPU计算无法作为独立的推理资源发挥作用,导致负载下CPU端进度波动加剧,并使尾部词间延迟(ITL)显著增加。我们提出了DualPace,一种将主机转变为独立调度解码引擎的LLM推理运行时。DualPace在两个引擎之间划分了调度权:GPU引擎

arXiv cs.DC

RSI-Opt:用于模型训练的递归自我改进优化代理

arXiv:2610.32026v1 发布类型:新 摘要:传统的神经网络训练通常遵循预先设定的优化方案,为此我们设计了一种LLM智能体,用于监控训练过程并随训练进展调整优化器参数;然而,固定不变的智能体策略无法在多次运行之间积累经验。我们提出了RSI-Opt,一种即插即用的递归自我改进优化器智能体,它能够不断演化指导这一过程的知识体系:包括一个提示、一套操作规程、一个经过验证的优化器代码库,以及一套决策节奏。一个元智能体基于先前运行中的证据重写这一策略,同时通过存档机制保留已评估的策略,以供后续进一步演化。在三项训练任务上历经五代迭代后,这些元智能体逐步形成了可复用的操作流程与代码,并自主开展其自身的…

arXiv cs.AI

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

翻译服务暂未运行,当前显示原文。
llama.cpp 发布