HappyZ happyz
happyz synced commits to refs/tags/b2729 at happyz/llama.cpp from mirror 2024-04-25 11:14:17 -07:00
happyz synced new reference refs/tags/b2728 to happyz/llama.cpp from mirror 2024-04-25 11:14:17 -07:00
happyz synced commits to refs/tags/b2728 at happyz/llama.cpp from mirror 2024-04-25 11:14:17 -07:00
happyz synced new reference refs/tags/b2727 to happyz/llama.cpp from mirror 2024-04-25 11:14:17 -07:00
happyz synced commits to refs/pull/6848/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:16 -07:00
ac829932a6 Increased opacity for contrast
fa0b4ad252 cmake : remove obsolete ANDROID check
d6e1d44f16 llama : synchronize before get/set session data (#6911)
853d06ffe2 ci : tmp disable slow tests
Compare 17 commits »
happyz synced commits to refs/pull/6866/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:16 -07:00
fa0b4ad252 cmake : remove obsolete ANDROID check
d6e1d44f16 llama : synchronize before get/set session data (#6911)
853d06ffe2 ci : tmp disable slow tests
3fe0596c18 readme : update model list (#6908)
Compare 12 commits »
happyz synced commits to refs/pull/6858/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:16 -07:00
ed7b3abb31 Merge 594e604de8ce7ee7ae8467f8ea5803b1b4a3498d into fa0b4ad252
fa0b4ad252 cmake : remove obsolete ANDROID check
d6e1d44f16 llama : synchronize before get/set session data (#6911)
853d06ffe2 ci : tmp disable slow tests
3fe0596c18 readme : update model list (#6908)
Compare 13 commits »
happyz synced commits to refs/pull/6858/head at happyz/llama.cpp from mirror 2024-04-25 11:14:16 -07:00
594e604de8 Allow params.js to be embedded in server.cpp
happyz synced commits to refs/pull/6844/head at happyz/llama.cpp from mirror 2024-04-25 11:14:15 -07:00
238551ed8c parse gmml_type and llama_ftype, allow specifiying cfg file
happyz synced commits to refs/pull/6848/head at happyz/llama.cpp from mirror 2024-04-25 11:14:15 -07:00
ac829932a6 Increased opacity for contrast
43daf2fe3c Trailing whitespace
c7c03260f6 Newline
ae2a08114d Newline
32a9792275 Newline
Compare 5 commits »
happyz synced commits to refs/pull/6840/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:15 -07:00
f195dd8710 Merge bc1585b44449135e289c94bcb97bed3bf5423dd8 into fa0b4ad252
fa0b4ad252 cmake : remove obsolete ANDROID check
d6e1d44f16 llama : synchronize before get/set session data (#6911)
853d06ffe2 ci : tmp disable slow tests
3fe0596c18 readme : update model list (#6908)
Compare 13 commits »
happyz synced commits to refs/pull/6840/head at happyz/llama.cpp from mirror 2024-04-25 11:14:15 -07:00
bc1585b444 llamafile : improve moe prompt eval speed on cpu
784e11dea1 README: add graphic for matrix multiplication (#6881)
b4e4b8a935 llama : add llama_get_pooling_type function (#6862)
3fe847b574 server : do not apply Markdown formatting in code sections (#6850)
37246b1031 common : revert showing control tokens by default for server (#6860)
Compare 10 commits »
happyz synced commits to refs/pull/6844/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:15 -07:00
238551ed8c parse gmml_type and llama_ftype, allow specifiying cfg file
784e11dea1 README: add graphic for matrix multiplication (#6881)
Compare 3 commits »
happyz synced commits to refs/pull/6834/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:14 -07:00
e256a140e5 Merge 19f18433cf5cc9de287cf11c9882dc76fdbd9f77 into 853d06ffe2
853d06ffe2 ci : tmp disable slow tests
3fe0596c18 readme : update model list (#6908)
0ead1f1072 llama : check that all the tensor data is in the model file (#6885)
19f18433cf ChatON+Main: Updates wrt detailed meta json
Compare 20 commits »
happyz synced commits to refs/pull/6832/head at happyz/llama.cpp from mirror 2024-04-25 11:14:14 -07:00
f9b42b8cd8 Added new options and some fixes
happyz synced commits to refs/pull/6832/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:14 -07:00
fa0b4ad252 cmake : remove obsolete ANDROID check
d6e1d44f16 llama : synchronize before get/set session data (#6911)
853d06ffe2 ci : tmp disable slow tests
3fe0596c18 readme : update model list (#6908)
Compare 13 commits »
happyz synced commits to refs/pull/6834/head at happyz/llama.cpp from mirror 2024-04-25 11:14:14 -07:00
19f18433cf ChatON+Main: Updates wrt detailed meta json
1495d039bc ChatON: Update to new detailed format wrt llama2 and llama3
1076f810be ChatON: Backup the current simple meta json file
78706bed3a ChatON: Keep compiler happy
4b02deed03 ChatON:MetaOK->MetaDump: Alert if user->end is needed or not
Compare 10 commits »
happyz synced commits to refs/pull/6839/head at happyz/llama.cpp from mirror 2024-04-25 11:14:14 -07:00
75c37ed817 fixed bug in dry sampler
99b77600f1 added dry sampler implementatin
b03b419598 Merge pull request #9 from ggerganov/gg/flash-attn
9ca869876e batched-bench : add fattn arg
c16a7c2688 metal : use F32 attention accumulators
Compare 211 commits »
happyz synced commits to refs/pull/6831/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:13 -07:00
fa0b4ad252 cmake : remove obsolete ANDROID check
d6e1d44f16 llama : synchronize before get/set session data (#6911)
853d06ffe2 ci : tmp disable slow tests
3fe0596c18 readme : update model list (#6908)
Compare 12 commits »
happyz synced commits to refs/pull/6811/merge at happyz/llama.cpp from mirror 2024-04-25 11:14:13 -07:00
853d06ffe2 ci : tmp disable slow tests
3fe0596c18 readme : update model list (#6908)
0ead1f1072 llama : check that all the tensor data is in the model file (#6885)
51543729ff ggml : fix redefinition of vaddvq_f32 for 32-bit ARM (#6906)
Compare 10 commits »