Commit Graph

5927 Commits

Author SHA1 Message Date
Younes B
4610ee2020 Update src/llama-vocab.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2025-07-08 13:55:00 +04:00
Younes B
f8d7c970a7 Update src/llama-arch.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2025-07-08 13:54:53 +04:00
ibrahimkhadraoui
c3c5d51c6a added hashes 2025-07-08 13:37:14 +04:00
ibrahimkhadraoui
7edf380090 Merge branch 'add-fh1-rebased' of https://github.com/tiiuae/llama.cpp-public into add-fh1-rebased 2025-07-08 13:29:54 +04:00
ibrahim khadraoui
90ddf2412a Update convert_hf_to_gguf.py
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2025-07-08 13:23:56 +04:00
ibrahim khadraoui
212edffd86 Update src/llama-arch.cpp
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2025-07-08 13:23:37 +04:00
ibrahim khadraoui
debf4e5dd5 Update src/llama-model.cpp
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2025-07-08 13:23:19 +04:00
ibrahim khadraoui
40058c043f Update src/llama-model.cpp
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2025-07-08 13:23:10 +04:00
ibrahim khadraoui
7fe1794cc3 Update src/llama-hparams.cpp
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2025-07-08 13:22:56 +04:00
ibrahimkhadraoui
9b92648302 flake8 fixes 2025-07-08 13:14:47 +04:00
Younes B
d28c31a90c Merge branch 'master' into add-fh1-rebased 2025-07-08 10:37:13 +02:00
Younes B
58e3866d02 Update src/llama-model.cpp
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2025-07-08 12:30:55 +04:00
Xuan-Son Nguyen
8f22dc0a53 model : add hunyuan moe (#14425)
* model : add hunyuan moe

* tokenizer ok

* fix tensor name

* cgraph init

* chat template

* wip

* almost working

* skip embed, fix bos

* cleanup

* yarn scaling

* cleanup

* correct rope type

* failed token fix

* ntk alpha freq_base

* tokenization working

* cleanup and pr changes

* vocab_size sanity check

* ntk alpha generic

* Update convert_hf_to_gguf.py

* Apply suggestions from code review

* fix regression

* fix style

---------

Co-authored-by: kooshi <1934337+kooshi@users.noreply.github.com>
b5843
2025-07-08 11:24:06 +03:00
ibrahimkhadraoui
52d1ef35ba Merge branch 'add-fh1-rebased' of https://github.com/tiiuae/llama.cpp-public into add-fh1-rebased 2025-07-08 11:46:02 +04:00
ibrahimkhadraoui
9a048d8de9 flake8 fixes 2025-07-08 11:45:58 +04:00
Jeff Bolz
53903ae6fa vulkan: increase timeout for CI (#14574) 2025-07-08 09:38:31 +02:00
younesbelkada
097df0ed85 remove final_norm 2025-07-08 11:26:04 +04:00
younesbelkada
adff470c8a more cleanups and fixed conversion 2025-07-08 11:19:38 +04:00
Georgi Gerganov
4d0dcd4a06 cuda : fix rope with partial rotation and non-cont src (#14580)
* cuda : fix rope non-cont

ggml-ci

* cont : fix multi-rope + add test

ggml-ci

* sycl : try fix

ggml-ci

* cont : fix sycl + clean-up cuda

ggml-ci
b5841
2025-07-08 10:15:21 +03:00
younesbelkada
823696bab1 remove unneeded attributes 2025-07-08 11:15:21 +04:00
ibrahimkhadraoui
2834a4ac10 clean 2025-07-08 11:00:30 +04:00
younesbelkada
4bc9e0ca89 tensor not required 2025-07-08 10:56:34 +04:00
ibrahimkhadraoui
f266d145fc added falcon-h1 2025-07-08 10:53:48 +04:00
ibrahimkhadraoui
d41f111462 Merge branch 'add-fh1-rebased' of https://github.com/tiiuae/llama.cpp-public into add-fh1-rebased 2025-07-08 10:48:07 +04:00
ibrahimkhadraoui
f028a43a91 Merge branch 'add-fh1-rebased' of https://github.com/tiiuae/llama.cpp-public into add-fh1-rebased 2025-07-08 10:48:01 +04:00
younesbelkada
a846d02327 remove todo 2025-07-08 10:44:59 +04:00
Younes B
2dee7cf964 Apply suggestions from code review
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2025-07-08 10:43:50 +04:00
ibrahimkhadraoui
7846c67e5c minor cleanups 2025-07-08 10:42:15 +04:00
younesbelkada
8555ee8b2c more cleanups on python conversion; 2025-07-08 10:41:33 +04:00
younesbelkada
d473d42832 more cleanups 2025-07-08 10:39:12 +04:00
ibrahimkhadraoui
e63ee4649e cleanup 2025-07-08 10:31:12 +04:00
ibrahimkhadraoui
da8a338531 Merge branch 'add-fh1-rebased' of https://github.com/tiiuae/llama.cpp-public into add-fh1-rebased 2025-07-08 10:23:18 +04:00
ibrahimkhadraoui
67b2664290 cleaning unused hparams 2025-07-08 10:20:17 +04:00
younesbelkada
7d7da0b37e d_ssm -> d_inner; 2025-07-08 10:18:43 +04:00
Aman Gupta
75c91de6e9 CUDA: add bilinear interpolation for upscale (#14563) b5840 2025-07-08 10:11:18 +08:00
R0CKSTAR
68155c66f0 musa: fix build warnings (unused variable) (#14561)
Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>
b5839
2025-07-08 07:58:30 +08:00
Sigbjørn Skjæret
e1a7059053 llama : fix incorrect minicpm3 v_states shape (#14571) b5838 2025-07-07 23:35:35 +02:00
Sigbjørn Skjæret
12f55c302b llama : remove ggml_cont where possible (#14568) b5837 2025-07-07 21:35:08 +02:00
Aman Gupta
b9c3eefde1 CUDA: add bf16 and i32 to getrows (#14529) b5836 2025-07-07 21:45:43 +08:00
younesbelkada
d2f46f18ac moe cleanuips 2025-07-07 17:36:22 +04:00
younesbelkada
68cb7845e9 more cleanups 2025-07-07 17:34:20 +04:00
Younes B
fd203302aa Update src/llama-model-loader.cpp 2025-07-07 17:29:50 +04:00
younesbelkada
084873c215 some cleanups 2025-07-07 17:28:08 +04:00
younesbelkada
632861e6c1 some cleanups 2025-07-07 17:27:34 +04:00
younesbelkada
f74e266f04 fix comment 2025-07-07 17:23:47 +04:00
ibrahimkhadraoui
042e5ff90b cleaning debug quant 2025-07-07 17:21:54 +04:00
ibrahimkhadraoui
624699c53f cleaning debugging stuff 2025-07-07 17:20:24 +04:00
ibrahimkhadraoui
935d46fab0 changed ROPE_TYPE 2025-07-07 17:01:54 +04:00
ibrahimkhadraoui
b6df0a49d5 add bos False 2025-07-07 16:57:52 +04:00
ibrahimkhadraoui
ae937f442c rm unused key 2025-07-07 16:57:36 +04:00