Tags

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

b2755

e00b4a8f · Fix more int overflow during quant (PPL/CUDA). (#6563) · Apr 29, 2024
b2754

7bb36ccf · gguf : enforce that tensor names are unique (#6905) · Apr 28, 2024
b2753

ce023f6f · add device version in device list (#6959) · Apr 28, 2024
b2751

4dba7e81 · Replace "alternative" boolean operator in conditional compilation directive (#6949) · Apr 27, 2024
b2750

b7368332 · ci: server: tests python env on github container ubuntu latest / fix n_predict (#6935) · Apr 27, 2024
b2749

928e0b70 · Reset schedule earlier to allow overlap with ggml graph computation on device (#6933) · Apr 26, 2024
b2748

0c4d489e · quantize: add imatrix and dataset metadata in GGUF (#6658) · Apr 26, 2024
b2747

017e6999 · add basic tensor data validation function (#6884) · Apr 26, 2024
b2746

e2764cd7 · gguf : fix mismatch between alloc and free functions (#6929) · Apr 26, 2024
b2740

d4a9afc1 · ci: server: fix python installation (#6918) · Apr 26, 2024
b2737

46e12c46 · llava : add support for moondream vision language model (#6899) · Apr 25, 2024
b2736

dba497e0 · cmake : restore LLAMA_LLAMAFILE_DEFAULT · Apr 25, 2024
b2735

fa0b4ad2 · cmake : remove obsolete ANDROID check · Apr 25, 2024
b2734

d6e1d44f · llama : synchronize before get/set session data (#6911) · Apr 25, 2024
b2731

0ead1f10 · llama : check that all the tensor data is in the model file (#6885) · Apr 25, 2024
b2730

51543729 · ggml : fix redefinition of vaddvq_f32 for 32-bit ARM (#6906) · Apr 25, 2024
b2729

4ab99d8d · clip : rename lerp function to avoid conflict (#6894) · Apr 25, 2024
b2728

54770413 · ggml : fix MIN / MAX macros (#6904) · Apr 25, 2024
b2727

aa750c1e · tests : minor bash stuff (#6902) · Apr 25, 2024
b2724

b4e4b8a9 · llama : add llama_get_pooling_type function (#6862) · Apr 24, 2024

Prev
1
…
8
9
10
11
12
13
14
15
16
…
99
Next