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

b4572

4bf3119d · cmake : don't fail on `GGML_CPU=OFF` (#11457) · Jan 28, 2025
b4570

6e84b0ab · SYCL : SOFTMAX F16 mask support and other fixes (#11261) · Jan 28, 2025
b4569

2b8525d5 · Handle missing model in CLI parameters for llama-run (#11399) · Jan 28, 2025
b4568

a4417ddd · Add new hf protocol for ollama (#11449) · Jan 27, 2025
b4567

d6d24cd9 · AMD: parse the architecture as supplied by gcnArchName (#11244) · Jan 27, 2025
b4566

a5203b44 · llama : minor fixes for up llama load model speed (#11448) · Jan 27, 2025
b4565

df984e01 · llama: refactor llama_decode_impl (#11381) · Jan 27, 2025
b4564

acd38efe · metal: Handle null returned from MTLCreateSystemDefaultDevice() (#11441) · Jan 27, 2025
b4562

178a7eb9 · metal : use residency sets (#11427) · Jan 26, 2025
b4560

19f65187 · cmake: add ggml find package (#11369) · Jan 26, 2025
b4559

1d8ee060 · rpc: fix register position (#11424) · Jan 26, 2025
b4557

f35726c2 · build: apply MSVC /bigobj option to c/cpp files only (#11423) · Jan 26, 2025
b4552

49b0e3ce · server : fix cleaning up stream task (#11418) · Jan 25, 2025
b4550

00c24acb · ci : fix line breaks on windows builds (#11409) · Jan 25, 2025
b4549

466ea66f · CANN: Add Ascend CANN build ci (#10217) · Jan 25, 2025
b4548

5f0db952 · hip : Add hipGraph and VMM support to ROCM (#11362) · Jan 25, 2025
b4547

c5d9effb · CUDA: fix FP16 cuBLAS GEMM (#11396) · Jan 24, 2025
b4546

9fbadaef · rocBLAS: Avoid fp32->fp16->fp32 conversion on cdna (#11356) · Jan 24, 2025
b4545

9755129c · release : pack /lib in the packages (#11392) · Jan 24, 2025
b4543

8137b4bb · CPU/CUDA: fix (GQA) mul mat back, add CUDA support (#11380) · Jan 24, 2025

1
…
25
26
27
28
29
30
31
32
33
…
178