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

b5018

a6f32f0b · Fix clang warning in gguf_check_reserved_keys (#12686) · Apr 01, 2025
b5017

2bb3597e · vulkan: fix build when glslc doesn't support coopmat (#12683) · Apr 01, 2025
b5016

82939705 · SYCL: Rename oneMKL to oneMath (#12192) · Apr 01, 2025
b5015

8bbf2608 · SYCL: switch to SYCL namespace (#12674) · Apr 01, 2025
b5013

c80a7759 · vocab : add special infill tokens for CodeLlama (#11850) · Mar 31, 2025
b5012

250d7953 · ggml : faster ssm scan (#10558) · Mar 31, 2025
b5010

a8a1f335 · Vulkan: Add DP4A MMQ and Q8_1 quantization shader (#12135) · Mar 31, 2025
b5009

1790e731 · cmake : fix whitespace (#0) · Mar 31, 2025
b5006

1a859490 · llava : proper description fix (#12668) · Mar 31, 2025
b5005

6c02a032 · SYCL: Remove misleading ggml_sycl_op_flatten function (#12387) · Mar 31, 2025
b5004

f52d59d7 · llava : fix clip loading GGUFs with missing description (#12660) · Mar 31, 2025
b5003

52de2e59 · tts : remove printfs (#12640) · Mar 31, 2025
b5002

2c3f8b85 · llama : support BailingMoE (Ling) (#12634) · Mar 30, 2025
b5001

4663bd35 · metal : use constexpr in FA kernels + fix typedef (#12659) · Mar 30, 2025
b4999

7242dd96 · llama-chat : Add Yandex instruct model template support (#12621) · Mar 30, 2025
b4998

492d7f1f · musa: fix all warnings, re-enable `-DLLAMA_FATAL_WARNINGS=ON` in ci and update doc (#12611) · Mar 30, 2025
b4997

d3f1f0ac · sync : ggml · Mar 30, 2025
b4992

af6ae1ef · llama : fix non-causal mask for gemma 3 (#12615) · Mar 30, 2025
b4991

0bb29193 · llama : change cpu_buft_list order: ACCEL -> GPU host -> CPU extra -> CPU (#12632) · Mar 29, 2025
b4990

a69f8463 · cmake : fix ccache conflict (#12522) · Mar 29, 2025

1
…
10
11
12
13
14
15
16
17
18
…
178