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

b4485

f446c2cf · SYCL: Add gated linear attention kernel (#11175) · Jan 15, 2025
b4481

091592d7 · Refactor test-chat-template.cpp (#11224) · Jan 14, 2025
b4475

84a44815 · cli : auto activate conversation mode if chat template is available (#11214) · Jan 13, 2025
b4474

39509fb0 · cuda : CUDA Graph Compute Function Refactor (precursor for performance improvements) (#11042) · Jan 13, 2025
b4468

8f70fc3d · llama : remove 'd' from bad special token log (#11212) · Jan 13, 2025
b4467

1244cdcf · ggml : do not define GGML_USE_CUDA when building with GGML_BACKEND_DL (#11211) · Jan 13, 2025
b4466

924518e2 · Reset color before we exit (#11205) · Jan 12, 2025
b4465

9a483999 · llama : fix chat template gguf key (#11201) · Jan 12, 2025
b4464

08f10f69 · llama : remove notion of CLS token (#11064) · Jan 12, 2025
b4458

c3f9d257 · Vulkan: Fix float16 use on devices without float16 support + fix... · Jan 10, 2025
b4457

ee7136c6 · llama: add support for QRWKV6 model architecture (#11001) · Jan 10, 2025
b4456

c6860cc7 · SYCL: Refactor ggml_sycl_compute_forward (#11121) · Jan 10, 2025
b4453

f8feb4b0 · model: Add support for PhiMoE arch (#11003) · Jan 09, 2025
b4451

d9feae1c · llama-chat : add phi 4 template (#11148) · Jan 09, 2025
b4450

8d59d911 · fix: add missing msg in static_assert (#11143) · Jan 08, 2025
gguf-v0.14.0

8a1d9c25 · gguf-py : move scripts directory (#11116) · Jan 08, 2025
b4447

f7cd1330 · ci : use actions from ggml-org (#11140) · Jan 08, 2025
b4446

4d2b3d88 · lora : improve compat with `mergekit-extract-lora` (#11131) · Jan 08, 2025
b4445

c07d437b · llama : avoid hardcoded QK_K (#11061) · Jan 08, 2025
b4443

c792dcf4 · ggml : allow loading backend with env variable (ggml/1059) · Jan 08, 2025

1
…
28
29
30
31
32
33
34
35
36
…
178