跳到正文
AI NEWS看见智能的每一步
资讯收藏
THE JOURNAL
AI NEWS JOURNAL
AI, IN PERSPECTIVE

全部 AI 动态

保留完整资讯流，用来源、主题和时间找到你关心的消息。

按发布时间排序

Transformers releases

版本：v5.21.0

# 发布版本 v5.21.0 ## 新增模型 ### MiniCPM-OCR MiniCPM-OCR 是 OpenBMB 的 MiniCPM-V 系列中的一款文档理解模型。它沿用了 MiniCPM-V 4.6 的架构，即由 SigLIP 视觉编码器与窗口注意力融合模块搭配 Qwen3.5 语言模型组成。该模型专为解析文本密集的图像而设计，例如扫描页、截图、收据和表格等。 **链接：** [文档](https://huggingface.co/docs/transformers/main/en/model_doc/minicpm_ocr) | [论文](https://huggingface.co/papers/2509.18154) * @guarin 在 [#49920](https://github.com/huggingface/transformers/pull/49920) 中提交的特性测试 (#49920) ### Flint Flint 是 Recursive 公司的专家混合语言模型。与 Qwen3-MoE 等标准的 SwiGLU MoE 变换器相比，它采用了混合注意力机制（滑动窗口…

事件进展与其他报道
llama.cpp releases

b11901

sycl：将临时内存的绑定大小调整为32 MiB（#31062）**网站：** - **证明文件：** - **macOS/iOS：** - [macOS Apple Silicon（arm64）](https://github.com/ggml-org/llama.cpp/releases/download/b11901/llama-b11901-bin-macos-arm64.tar.gz) - macOS Apple Silicon（arm64，已启用KleidiAI）[已禁用](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel（x64）](https://github.com/ggml-org/llama.cpp/releases/download/b11901/llama-b11901-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11901/llama-b11901-xcframework.zip) **Linux：** - [Ubuntu x64（CPU）](https://github.com/ggml-org/llama.cpp/releases/download/b11901/llama-b11901-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64（CPU）](https://github.com/ggml-org/llama.cpp/releases/download/b11901/llama-b)

事件进展与其他报道

llama.cpp releases

b11900

common：在真实的 HF 代码仓库布局上固定模型解析结果（#30548）* tests：在真实的仓库列表上固定模型解析结果 通过从 HF API 原封不动地抓取的 63 个仓库列表扩展 test-model-resolution 测试，这些列表经过精心挑选，旨在覆盖主要出版商的 1237 个 GGUF 仓库中出现的所有文件命名模式，并进一步筛选为整数量化组。每个仓库都会固定其解析后的计划以及缓存列表。预期结果依据的是每个文件在其 GGUF 头部中实际所代表的内容，而非当前解析器的判断，因此一旦出现不匹配即视为解析错误。不匹配项会被统一收集并报告，随后测试才会失败。* tests：对于文件系统拒绝访问的路径，保持列表状态未检查 一些真实的拆分名称（例如 DevQuasar 的 Model.Q2_K.gguf-00001-of-00012.gguf 文件位于某个量化文件夹下）在实际展开后会超出 MAX_PATH 的限制

事件进展与其他报道
llama.cpp releases

b11899

mtmd：添加 mtmd_helper_get_capabilities（#31183）* mtmd：添加 mtmd_helper_get_capabilities * 支持 audio_stream * 为保持兼容性恢复旧版 API * audio_gen_spk_ref、text_gen **网站：** - **证明文件：** - **macOS/iOS：** - [macOS Apple Silicon（arm64）](https://github.com/ggml-org/llama.cpp/releases/download/b11899/llama-b11899-bin-macos-arm64.tar.gz) - macOS Apple Silicon（arm64，已启用 KleidiAI）[已禁用](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel（x64）](https://github.com/ggml-org/llama.cpp/releases/download/b11899/llama-b11899-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11899/llama-b11899-xcframework.zip) **Linux：** - [Ubuntu x64（CPU）](https://github.com/ggml-org/llama.cpp/releases/download/b11899/llama-b11899-)

事件进展与其他报道
llama.cpp releases

b11898

ggml-cpu : fix unused variables in arm arch (#31180) **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11898/llama-b11898-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/b11898/llama-b11898-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11898/llama-b11898-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11898/llama-b11898-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11898/llama-

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

b11897

tests : refactor test-llama-archs to not link internal libllama symbols (#30880) test-llama-archs included src/llama-arch.h and src/llama-model-saver.h, which prevented building on Windows with shared libraries since the used symbols are not exported from the DLL. - declare llama_arch_* wrappers around the internal llm_ arch query functions in the llama-ext.h staging header, together with llama_arch_kv_name() and llama_arch_supports_save() - slim llama_model_saver down to the private implementation of llama_model_save_to_file(): the fixture-writer half used by the test moves into tests/test-llama-archs.cpp as a local kv_writer - replace the roundtrip saver usage with the public llama_model_save_to_file() writing to a temp path (also removes the tmpfile() fd leak and its Windows workaround)

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

b11896

ci : fix race in webgpu memset (#31162) **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11896/llama-b11896-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/b11896/llama-b11896-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11896/llama-b11896-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11896/llama-b11896-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11896/llama-b11896-bin-ub

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

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

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

b11890

deepseekv4.1: follow-up review fixes (#31137) * kv-cache : use ggml_backend_tensor_copy to refill the DSV4 K mirrors * llama : move the Engram pad id to llama_hparams and the token map to llama_model The saver no longer needs the deepseek41 model class. * deepseek41 : drop the F32 src1 request on the HC mixes matmul * ggml : add softplus TODOs in the backends that still use log(1 + exp(x)) * convert : make the DSV4/DSV4.1 modify_tensors generators, drop the redundant vision filter * deepseek41 : clean up comments * deepseek41 : drop the F32 src1 requests on the router and indexer matmuls * convert : use format_tensor_name directly in DeepseekV4Model * s/comp_kv_src/comp_il_src/g * cont : fix TODOs for log1p --------- Co-authored-by: Georgi Gerganov **Website:** - **Attestations:** - **macO

译文排队中，当前为原文 · en
事件进展与其他报道
n8n releases

beta

## [2.45.4](https://github.com/n8n-io/n8n/compare/n8n@2.45.3...n8n@2.45.4) (2026-10-26) ### Bug Fixes * **core:** Recreate removed n8n-sandbox sandboxes and rerun workspace setup ([#41919](https://github.com/n8n-io/n8n/issues/41919)) ([11e279a](https://github.com/n8n-io/n8n/commit/11e279a51d497f8d96f5f9f90bb314dda4f6a49e)) * **core:** Start sub-workflow error workflow when a pre-run check fails ([#41977](https://github.com/n8n-io/n8n/issues/41977)) ([9595142](https://github.com/n8n-io/n8n/commit/95951428c765f3f0b1da9f341a6ab8c75673385a)) * **editor:** Match navigation plus button hover styling ([#41910](https://github.com/n8n-io/n8n/issues/41910)) ([471c006](https://github.com/n8n-io/n8n/commit/471c006fdd46091659f3e1d28d989702049489f0)) * **editor:** Point migration report docs link to the

译文排队中，当前为原文 · en
事件进展与其他报道
n8n releases

n8n@2.45.4

## [2.45.4](https://github.com/n8n-io/n8n/compare/n8n@2.45.3...n8n@2.45.4) (2026-10-26) ### Bug Fixes * **core:** Recreate removed n8n-sandbox sandboxes and rerun workspace setup ([#41919](https://github.com/n8n-io/n8n/issues/41919)) ([11e279a](https://github.com/n8n-io/n8n/commit/11e279a51d497f8d96f5f9f90bb314dda4f6a49e)) * **core:** Start sub-workflow error workflow when a pre-run check fails ([#41977](https://github.com/n8n-io/n8n/issues/41977)) ([9595142](https://github.com/n8n-io/n8n/commit/95951428c765f3f0b1da9f341a6ab8c75673385a)) * **editor:** Match navigation plus button hover styling ([#41910](https://github.com/n8n-io/n8n/issues/41910)) ([471c006](https://github.com/n8n-io/n8n/commit/471c006fdd46091659f3e1d28d989702049489f0)) * **editor:** Point migration report docs link to the

译文排队中，当前为原文 · en
事件进展与其他报道
n8n releases

stable

## [2.44.7](https://github.com/n8n-io/n8n/compare/n8n@2.44.6...n8n@2.44.7) (2026-10-26) ### Bug Fixes * **core:** Start sub-workflow error workflow when a pre-run check fails ([#41978](https://github.com/n8n-io/n8n/issues/41978)) ([67d0e65](https://github.com/n8n-io/n8n/commit/67d0e65f5d5d01d49b1d14ea2c26a6b803bf058b)) * **Email Trigger (IMAP) Node:** Back off reconnects when the server drops each new connection ([#41927](https://github.com/n8n-io/n8n/issues/41927)) ([10a9518](https://github.com/n8n-io/n8n/commit/10a9518f0d9d470e619b500383b5b76fc1870662)) * **Microsoft SharePoint Node:** Fix the column mapper 403 and empty items in Map Automatically ([#41906](https://github.com/n8n-io/n8n/issues/41906)) ([e67d278](https://github.com/n8n-io/n8n/commit/e67d278ef0ae3c15493e74813ecce972e384ff1

译文排队中，当前为原文 · en
事件进展与其他报道
n8n releases

n8n@2.44.7

## [2.44.7](https://github.com/n8n-io/n8n/compare/n8n@2.44.6...n8n@2.44.7) (2026-10-26) ### Bug Fixes * **core:** Start sub-workflow error workflow when a pre-run check fails ([#41978](https://github.com/n8n-io/n8n/issues/41978)) ([67d0e65](https://github.com/n8n-io/n8n/commit/67d0e65f5d5d01d49b1d14ea2c26a6b803bf058b)) * **Email Trigger (IMAP) Node:** Back off reconnects when the server drops each new connection ([#41927](https://github.com/n8n-io/n8n/issues/41927)) ([10a9518](https://github.com/n8n-io/n8n/commit/10a9518f0d9d470e619b500383b5b76fc1870662)) * **Microsoft SharePoint Node:** Fix the column mapper 403 and empty items in Map Automatically ([#41906](https://github.com/n8n-io/n8n/issues/41906)) ([e67d278](https://github.com/n8n-io/n8n/commit/e67d278ef0ae3c15493e74813ecce972e384ff1

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

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

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

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

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

b11887

tests : adjust default threads for test-llama-context (#31157) **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11887/llama-b11887-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/b11887/llama-b11887-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11887/llama-b11887-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11887/llama-b11887-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

b11886

qwen: merge tensors on convert (#31028) * qwen: merge tensors on convert * convert: use generator functions * qwen4exp: remove conts and pre-scale **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11886/llama-b11886-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/b11886/llama-b11886-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11886/llama-b11886-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11886/llama-b11886-bin-ubuntu-x64.tar.gz

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

b11884

llama: fix exhausting CPU and MEM caused by dry sequence breaker (#31035) **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11884/llama-b11884-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/b11884/llama-b11884-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11884/llama-b11884-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11884/llama-b11884-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/d

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

b11883

llama-quant : fix 4D tensor quantization using ggml_nrows() and name-based expert detection (#30820) * llama-quant : fix quantizing 4D tensors nrows_total was ne[1]*ne[2], so for a 4D tensor only the first ne[2] slices were quantized and the rest of the rows were dropped. Use ggml_nrows(tensor) to count all rows, including ne[3]. Fixes #29916 * llama-quant : detect expert tensors by name and fix imatrix size for 4D tensors The MXFP4_MOE path used ne[2] > 1 to detect MoE expert tensors. This is wrong for non-expert tensors with 3 or more dims, e.g. MLA projections (BailingMoE3) and 4D conv kernels. Use the tensor name instead: all tensors used by GGML_OP_MUL_MAT_ID are named *_exps or *_chexps. This also removes the BailingMoE3 special case. The imatrix holds one vector of ne[0] values per

译文排队中，当前为原文 · en
事件进展与其他报道
llama.cpp releases

b11882

grammar: limit recursion depth to prevent stack overflow (#29495) (#29755) **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11882/llama-b11882-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/b11882/llama-b11882-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11882/llama-b11882-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11882/llama-b11882-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/

译文排队中，当前为原文 · en
事件进展与其他报道