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

b4440

8cef75c7 · llamafile : ppc64le MMA INT8 implementation (#10912) · Jan 08, 2025
b4439

0d52a69e · ci : fix cmake option (#11125) · Jan 08, 2025
b4438

02f04301 · Disable GL_KHR_cooperative_matrix Vulkan extension if not available. (#11117) · Jan 08, 2025
b4437

bec2183f · fix: Vulkan shader gen binary path when Cross-compiling (#11096) · Jan 08, 2025
b4435

017cc5f4 · ggml-backend : only offload from host buffers (fix) (#11124) · Jan 07, 2025
b4434

a3d50bc0 · ggml-backend : only offload from host buffers (#11120) · Jan 07, 2025
b4433

a4dd4900 · rpc : code cleanup (#11107) · Jan 07, 2025
b4432

c0d6f790 · SYCL: Use get_multi_ptr instead of deprecated get_pointer in wkv6 (#11087) · Jan 07, 2025
b4431

dc7cef9f · llama-run : fix context size (#11094) · Jan 06, 2025
b4430

ecebbd29 · llama : remove unused headers (#11109) · Jan 06, 2025
b4428

e6e7c75d · server : fix extra BOS in infill endpoint (#11106) · Jan 06, 2025
b4426

96a1dc27 · llama : prevent system info string accumulation across calls (#11101) · Jan 06, 2025
b4425

6369f867 · llama : rename missed batch params/vars to ubatch (#10059) · Jan 06, 2025
b4424

47182dd0 · llama : update llama_model API names (#11063) · Jan 06, 2025
b4423

3e6e7a6b · tokenize : escape the prompt (#11058) · Jan 06, 2025
b4422

ae2f606b · mmap : fix fileno macro clash (#11076) · Jan 06, 2025
b4421

727368c6 · llama : use LLAMA_TOKEN_NULL (#11062) · Jan 06, 2025
b4420

5047dd35 · llama : use _impl suffix instead of _internal (#11060) · Jan 06, 2025
b4419

46e3556e · CUDA: add BF16 support (#11093) · Jan 06, 2025
b4418

b56f079e · Vulkan: Add device-specific blacklist for coopmat for the AMD proprietary driver (#11074) · Jan 04, 2025

1
…
29
30
31
32
33
34
35
36
37
…
178