Tags

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

b2371

8a3012a4 · ggml : add ggml-common.h to deduplicate shared code (#5940) · Mar 09, 2024
b2370

9674aaf3 · server : simplify logic for empty prompts (#5953) · Mar 09, 2024
b2369

950ba1ab · Server: reorganize some http logic (#5939) · Mar 09, 2024
b2368

e1fa9569 · server : add SSL support (#5926) · Mar 09, 2024
b2367

fd72d2d2 · server: tests: add truncated prompt tests, better kv cache size (#5933) · Mar 09, 2024
b2366

c2101a2e · llama : support Mamba Selective State Space Models (#5328) · Mar 08, 2024
b2365

515f7d0d · llama : fix quantization of shared token_embd (#5944) · Mar 08, 2024
b2364

76e86882 · server: metrics: add llamacpp:prompt_seconds_total and... · Mar 08, 2024
b2363

e457fb35 · llama : assume tied weights if lm_head/output weights is missing (#5824) · Mar 08, 2024
b2362

af37fd8b · server : fix EOS token detection with disabled cache (#5938) · Mar 08, 2024
b2361

581ed5c4 · log : fix MSVC compile errors (#5643) · Mar 08, 2024
b2360

6cdabe65 · llama-bench : add embeddings option (#5924) · Mar 07, 2024
b2359

89fb735f · Revert "[SYCL] fix error when set main gpu to non-zero (#5901)" (#5918) · Mar 07, 2024
b2358

55a2a900 · server : add `/v1/completions` endpoint (#5914) · Mar 07, 2024
b2357

2002bc96 · server : refactor (#5882) · Mar 07, 2024
b2356

ceca1aef · [SYCL] fix error when set main gpu to non-zero (#5901) · Mar 07, 2024
b2355

e04e04f8 · ggml : use SYS_get_cpu if SYS_getcpu is not defined (#5906) · Mar 06, 2024
b2354

e25fb4b1 · ggml : use `uint8x16_t` return type for `ggml_vqtbl1q_u8` (#5894) · Mar 06, 2024
b2352

8ced9f7e · add wait() to make code stable (#5895) · Mar 06, 2024
b2350

bd836944 · quants : use MM256_SET_M128I consistently to fix gcc 7 build (#5889) · Mar 05, 2024

Prev
1
…
18
19
20
21
22
23
24
25
26
…
99
Next