Tags

Tags give the ability to mark specific points in history as being important

b1596

37c746d6 · llama : add Qwen support (#4281) · Dec 01, 2023
b1595

880f5797 · llama : fix integer overflow during quantization (#4284) · Dec 01, 2023
b1593

ef47ec18 · ggml : add ggml_soft_max_ext (#4256) · Dec 01, 2023
b1592

1d144112 · server : add --log-disable to disable logging to file (#4260) · Dec 01, 2023
b1591

f43f0936 · server : add single-client multi-prompt support (#4232) · Dec 01, 2023
b1590

d2809a3b · make : fix Apple clang determination bug (#4272) · Dec 01, 2023
b1589

15f5d960 · build : fix build info generation and cleanup Makefile (#3920) · Dec 01, 2023
b1587

8efa0f6e · main : pass LOG_TEE callback to llama.cpp log (#4033) · Nov 30, 2023
b1583

f7f9e062 · cmake : fix the metal file foder path (#4217) · Nov 30, 2023
b1581

b18c66ca · llama : fix alignment of general.name in print meta (#4254) · Nov 30, 2023
b1579

954e2285 · llama : fix typical sampling (#4261) · Nov 30, 2023
b1575

64e64aa2 · ggml : restore abort() in GGML_ASSERT (#4242) · Nov 28, 2023
b1574

8406b092 · ggml : re-enable BLAS for CPU when src0 != F32 + remove redundant full offload... · Nov 28, 2023
b1573

b38a16df · cmake : fix issue with version info not getting baked into LlamaConfig.cmake (#3970) · Nov 27, 2023
b1571

bb03290c · examples : iOS example with swift ui (#4159) · Nov 27, 2023
b1570

f3b26981 · ggml : fix -Warray-bounds warning with gcc (#4231) · Nov 26, 2023
b1569

3e73d31d · lookahead : support `-n -1` infinite generation · Nov 26, 2023
b1567

922754a8 · lookahead : add example for lookahead decoding (#4207) · Nov 26, 2023
b1566

22da0553 · metal : fix yarn (#4220) · Nov 26, 2023
b1564

f837c3a9 · llama : grammar `reserve` space in `decode_utf8` (#4210) · Nov 25, 2023

Prev
1
…
43
44
45
46
47
48
49
50
51
…
99
Next