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

b3570

4134999e · gguf-py : Numpy dequantization for most types (#8939) · Aug 11, 2024
b3568

a21c6fd4 · update guide (#8909) · Aug 11, 2024
b3567

33309f66 · llama : check all graph nodes when searching for result_embd_pooled (#8956) · Aug 11, 2024
b3566

7c5bfd57 · Optimize Vulkan backend for better CPU performance and less GPU synchronization overhead. (#8943) · Aug 11, 2024
b3565

6e02327e · metal : fix uninitialized abort_callback (#8968) · Aug 10, 2024
b3564

7eb23840 · llama : default n_swa for phi-3 (#8931) · Aug 10, 2024
b3563

7c3f55c1 · Add support for encoder-only T5 models (#8900) · Aug 10, 2024
b3561

b72942fa · Merge commit from fork · Aug 09, 2024
b3560

6afd1a99 · llama : add support for lora adapters in T5 model (#8938) · Aug 09, 2024
b3559

272e3bd9 · make : fix llava obj file race (#8946) · Aug 09, 2024
b3557

3071c0a5 · llava : support MiniCPM-V-2.5 (#7599) · Aug 09, 2024
b3556

4305b57c · sync : ggml · Aug 09, 2024
b3551

345a686d · llama : reduce useless copies when saving session (#8916) · Aug 08, 2024
b3547

e44a561a · sync : ggml · Aug 08, 2024
b3543

ebd541a5 · make : clean llamafile objects (#8923) · Aug 08, 2024
b3542

15fa07a5 · make : use C compiler to build metal embed object (#8899) · Aug 07, 2024
b3541

be55695e · ggml-backend : fix async copy from CPU (#8897) · Aug 07, 2024
b3540

0478174d · [SYCL] Updated SYCL device filtering (#8901) · Aug 07, 2024
b3539

a8dbc6f7 · CUDA/HIP: fix tests/test-backend-ops (#8896) · Aug 07, 2024
b3538

506122d8 · llama-bench : add support for getting cpu info on Windows (#8824) · Aug 07, 2024

1
…
59
60
61
62
63
64
65
66
67
…
178