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

b3180

91c188d6 · Only use FIM middle token if it exists (#7648) · Jun 18, 2024
b3179

84f6de17 · Fix no gcc pragma on Windows (#7751) · Jun 18, 2024
b3178

61665277 · Allow compiling with CUDA without CUDA runtime installed (#7989) · Jun 18, 2024
b3177

b96f9afb · chore: clean useless beam search param (#7985) · Jun 18, 2024
b3175

5326bcce · ggml : sync · Jun 18, 2024
b3173

a94e6ff8 · update: support Qwen2-57B-A14B (#7835) · Jun 17, 2024
b3172

5b6da187 · Make updates to type cast based on compiler instead of OS (#7851) · Jun 17, 2024
b3171

7c26775a · llama : disable FA if KV head size do not match (#7982) · Jun 17, 2024
b3169

99052cd2 · sched : offload_op also requires supports_op (#7977) · Jun 17, 2024
b3168

c637fcd3 · fix: divide 0 exception in mamba (#7932) · Jun 17, 2024
b3167

6a2f0b34 · Implement non-mapped async IO for CUDA on Windows. (#7896) · Jun 17, 2024
b3166

21be9cab · rpc : fix load/store misaligned addresses (#7948) · Jun 17, 2024
b3164

df68d4fa · [SYCL] Update README-sycl.md for Chapter "Recommended release" and "News" (#7946) · Jun 17, 2024
b3163

43b35e38 · Add support for sqrt on CUDA (#7953) · Jun 17, 2024
b3162

19b7a836 · cuda : fix bounds check for src0 rows in MMVQ kernel (whisper/2231) · Jun 16, 2024
b3159

bc6c457f · flake.lock: Update (#7951) · Jun 16, 2024
b3158

52399254 · unicode : avoid char32_t (#7957) · Jun 16, 2024
b3156

cddaf028 · ggml : fix handling of zero blocks in IQ quants (#7955) · Jun 16, 2024
b3154

7c7836d9 · Vulkan Shader Refactor, Memory Debugging Option (#7947) · Jun 16, 2024
b3153

0c7b3595 · Add `cvector-generator` example (#7514) · Jun 15, 2024

1
…
72
73
74
75
76
77
78
79
80
…
178