为了安全，强烈建议开启2FA双因子认证：User Settings -> Account -> Enable two-factor authentication！！！

Tags

Tags give the ability to mark specific points in history as being important

b4759

f3e64859 · ci : fix arm upload artifacts (#12024) · Feb 22, 2025
gg-ci-fix-arm-b4760-f343850

f343850b · cont : fix archive name to use matrix · Feb 22, 2025
b4756

cf756d6e · server : disable Nagle's algorithm (#12020) · Feb 22, 2025
b4755

d7090842 · cuda: Add Q5_1, Q5_0, Q4_1 and Q4_0 to F32 conversion support. (#12000) · Feb 22, 2025
b4754

de8b5a36 · llama.swiftui : add "Done" dismiss button to help view (#11998) · Feb 22, 2025
b4753

51f311e0 · llama : skip loading unused tensors (#12004) · Feb 21, 2025
b4751

ecc8e3ae · CUDA: correct the lowest Maxwell supported by CUDA 12 (#11984) · Feb 21, 2025
b4749

ee02ad02 · clip : fix visual encoders with no CLS (#11982) · Feb 21, 2025
b4747

c5d91a74 · ggml-cpu: Add CPU backend support for KleidiAI library (#11390) · Feb 20, 2025
b4746

4806498b · ggml: aarch64: implement SVE kernels for q3_K_q8_K vector dot (#11917) · Feb 20, 2025
b4745

0d559580 · run : add --chat-template-file (#11961) · Feb 20, 2025
b4743

d07c6213 · common : add llama.vim preset for Qwen2.5 Coder (#11945) · Feb 19, 2025
b4742

abd4d0bc · speculative : update default params (#11954) · Feb 19, 2025
b4739

63e489c0 · tool-call: refactor common chat / tool-call api (+ tests / fixes) (#11900) · Feb 18, 2025
b4738

63ac1285 · server : add TEI API format for /rerank endpoint (#11942) · Feb 18, 2025
b4735

73e2ed3c · CUDA: use async data loading for FlashAttention (#11894) · Feb 17, 2025
b4734

f7b1116a · update release requirements (#11897) · Feb 17, 2025
b4733

c4d29baf · server : fix divide-by-zero in metrics reporting (#11915) · Feb 17, 2025
b4732

2eea03d8 · vulkan: implement several ops relevant for ggml_opt (#11769) · Feb 17, 2025
b4731

0f2bbe65 · server : bump httplib to 0.19.0 (#11908) · Feb 16, 2025

1
…
19
20
21
22
23
24
25
26
27
…
178