 Georgi Gerganov
|
afa8a9ec9b
llama : add `llama_vocab`, functions -> methods, naming (#11110)
|
il y a 1 an |
 Johannes Gäßler
|
53ff6b9b9f
GGUF: C++ refactor, backend support, misc fixes (#11030)
|
il y a 1 an |
|
Georgi Gerganov
|
47182dd03f
llama : update llama_model API names (#11063)
|
il y a 1 an |
|
Georgi Gerganov
|
727368c60f
llama : use LLAMA_TOKEN_NULL (#11062)
|
il y a 1 an |
 Molly Sophia
|
4b0c638b9a
common : disable KV cache shifting automatically for unsupported models (#11053)
|
il y a 1 an |
|
Georgi Gerganov
|
f66f582927
llama : refactor `src/llama.cpp` (#10902)
|
il y a 1 an |
 Xuan Son Nguyen
|
45095a61bf
server : clean up built-in template detection (#11026)
|
il y a 1 an |
 Peter
|
6e1531aca5
common, examples, ggml : fix MSYS2 GCC compiler errors and warnings when building with LLAMA_CURL=ON and GGML_OPENCL=ON (#11013)
|
il y a 1 an |
|
Georgi Gerganov
|
0bf2d10c55
tts : add OuteTTS support (#10784)
|
il y a 1 an |
|
Georgi Gerganov
|
152610eda9
server : output embeddings for all tokens when pooling = none (#10861)
|
il y a 1 an |
|
Georgi Gerganov
|
644fd71b44
sampling : refactor + optimize penalties sampler (#10803)
|
il y a 1 an |
 Eric Curtin
|
c27ac678dd
Opt class for positional argument handling (#10508)
|
il y a 1 an |
|
Xuan Son Nguyen
|
adffa6ffd5
common : improve -ctv -ctk CLI arguments (#10806)
|
il y a 1 an |
 Diego Devesa
|
7cc2d2c889
ggml : move AMX to the CPU backend (#10570)
|
il y a 1 an |
|
Xuan Son Nguyen
|
9f912511bc
common : fix duplicated file name with hf_repo and hf_file (#10550)
|
il y a 1 an |
|
Diego Devesa
|
10bce0450f
llama : accept a list of devices to use to offload a model (#10497)
|
il y a 1 an |
|
Diego Devesa
|
5931c1f233
ggml : add support for dynamic loading of backends (#10469)
|
il y a 1 an |
|
Georgi Gerganov
|
d9d54e498d
speculative : refactor and add a simpler example (#10362)
|
il y a 1 an |
|
Georgi Gerganov
|
8e752a777b
llama : add check for KV cache shifts (#10401)
|
il y a 1 an |
|
Johannes Gäßler
|
4e54be0ec6
llama/ex: remove --logdir argument (#10339)
|
il y a 1 an |
|
Diego Devesa
|
ae8de6d50a
ggml : build backends as libraries (#10256)
|
il y a 1 an |
|
Georgi Gerganov
|
5c333e0140
metal : add BF16 support (#8439)
|
il y a 1 an |
|
Diego Devesa
|
9f40989351
ggml : move CPU backend to a separate file (#10144)
|
il y a 1 an |
|
Georgi Gerganov
|
8d8ff71536
llama : remove Tail-Free sampling (#10071)
|
il y a 1 an |
 wwoodsTM
|
ff252ea48e
llama : add DRY sampler (#9702)
|
il y a 1 an |
 Michael Podvitskiy
|
d80fb71f8b
llama: string_split fix (#10022)
|
il y a 1 an |
|
Georgi Gerganov
|
f594bc80ba
ggml : add asserts for type conversion in fattn kernels (#9971)
|
il y a 1 an |
|
Xuan Son Nguyen
|
cda0e4b648
llama : remove all_pos_0, all_pos_1, all_seq_id from llama_batch (#9745)
|
il y a 1 an |
 MaggotHATE
|
fbc98b748e
sampling : add XTC sampler (#9742)
|
il y a 1 an |
|
Georgi Gerganov
|
11ac9800af
llama : improve infill support and special token detection (#9798)
|
il y a 1 an |
