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

b3240

f2d48fff · sync : ggml · Jun 26, 2024
b3233

494165f3 · llama : extend llm_build_ffn() to support _scale tensors (#8103) · Jun 26, 2024
b3232

9b2f16f8 · `json`: better support for "type" unions (e.g. nullable arrays w/ typed items) (#7863) · Jun 26, 2024
b3231

6777c544 · `json`: fix additionalProperties, allow space after enum/const (#7840) · Jun 26, 2024
b3230

163d50ad · fixes #7999 (adds control vectors to all `build_XXX()` functions in... · Jun 25, 2024
b3229

6fcbf682 · llama : implement Unigram tokenizer needed by T5 and FLAN-T5 model families (#5763) · Jun 25, 2024
b3228

e6bf0077 · llama : return nullptr from llama_grammar_init (#8093) · Jun 25, 2024
b3227

84631fe1 · `json`: support integer minimum, maximum, exclusiveMinimum, exclusiveMaximum (#7797) · Jun 25, 2024
b3226

dd047b47 · disable docker CI on pull requests (#8110) · Jun 25, 2024
b3223

49c03c79 · cvector: better prompt handling, add "mean vector" method (#8069) · Jun 25, 2024
b3222

48e6b92c · Add chat template support for llama-cli (#8068) · Jun 25, 2024
b3220

f702a90e · Update control vector help (#8104) · Jun 25, 2024
b3219

083bacce · [SYCL] Re-enabled mul_mat_batched_sycl (#8095) · Jun 25, 2024
b3218

2df373ac · CUDA: fix matrix multiplication algorithm choice (#8102) · Jun 25, 2024
b3216

a818f302 · CUDA: use MMQ instead of cuBLAS by default (#8075) · Jun 24, 2024
b3212

8cb508d0 · disable publishing the full-rocm docker image (#8083) · Jun 24, 2024
b3211

646ef4a9 · embedding : more cli arguments (#7458) · Jun 24, 2024
b3209

95f57bb5 · ggml : remove ggml_task_type and GGML_PERF (#8017) · Jun 24, 2024
b3208

e112b610 · llama : add support for BitnetForCausalLM (#7931) · Jun 23, 2024
b3206

11318d9a · Fix typo in llama_set_embeddings comment (#8077) · Jun 23, 2024

1
…
70
71
72
73
74
75
76
77
78
…
178