Alex Cheema
|
3d0e2f1d0c
fix preemptive downloads with ensure_shard
|
9 сар өмнө |
Alex Cheema
|
f2d5beee08
change chatgpt api port from 8000 to 52415
|
9 сар өмнө |
Caden MacKenzie
|
bd2985aebd
Merge pull request #2 from cadenmackenzie/downloadedModelsV2Revisions
|
9 сар өмнө |
cadenmackenzie
|
dd38924e39
removing checking of percentage for models that are not found locally
|
9 сар өмнө |
cadenmackenzie
|
972074e98b
reducing redundent checks
|
9 сар өмнө |
cadenmackenzie
|
dfcf513d55
removing is_model_downloaded method and changing how downloaded variable is set
|
9 сар өмнө |
cadenmackenzie
|
d9aabd7802
working versions
|
9 сар өмнө |
Alex Cheema
|
f1eec9fa64
qwen-2.5-0.5b
|
9 сар өмнө |
Alex Cheema
|
fd867256e0
healthcheck
|
9 сар өмнө |
Alex Cheema
|
cafc6d37dd
Merge pull request #454 from giuseppegambino92/configure-mlx_dinamically
|
9 сар өмнө |
Caden MacKenzie
|
372d873fd0
Merge pull request #1 from dtnewman/dn/downloadModelsV2
|
9 сар өмнө |
Daniel Newman
|
cbeb1b33c2
fix safari issue
|
9 сар өмнө |
cadenmackenzie
|
3eb726cee0
removing sorting of models by name
|
9 сар өмнө |
cadenmackenzie
|
95ce665758
removing unneccesary css
|
9 сар өмнө |
cadenmackenzie
|
25d67f5096
cleaning up logging in index.js
|
9 сар өмнө |
Giuseppe Gambino
|
84ce076860
Edit configure_mlx.sh for calculate dinamically the value for iogpu.wired_limit_mb and iogpu.wired_lwm_mb. The script limit wired_limit_mb to 80% and wired_lwm_mb to 70%, but this threshold are variables.
|
9 сар өмнө |
cadenmackenzie
|
59f5b6d845
adding back in set error message
|
9 сар өмнө |
cadenmackenzie
|
fb32a851b1
removing error separtation so I can put in different PR
|
9 сар өмнө |
cadenmackenzie
|
7d7bdd83ed
removing uneccesary console logs and fixing order of variables in index.js
|
9 сар өмнө |
cadenmackenzie
|
de09e2a831
reusing helper function to get cached directory
|
9 сар өмнө |
cadenmackenzie
|
c7dd3126b1
adding logic to check which models are downloaded
|
9 сар өмнө |
Giuseppe Gambino
|
b0d7c34efa
Edit configure_mlx.sh for calculate dinamically the value for iogpu.wired_limit_mb and iogpu.wired_lwm_mb. The script limit wired_limit_mb to 80% and wired_lwm_mb to 60%, but this threshold are variables.
|
9 сар өмнө |
Alex Cheema
|
b6945224fa
disable configure_mlx.sh for now
|
9 сар өмнө |
Alex Cheema
|
34f3c4a155
fix tokenizers test with restructured models
|
9 сар өмнө |
Alex Cheema
|
65d2ae0287
Merge pull request #447 from cadenmackenzie/errorMessageTimeout
|
9 сар өмнө |
Alex Cheema
|
bfdad19d8e
Merge pull request #448 from blindcrone/grpc_compile_script
|
9 сар өмнө |
Alex Cheema
|
7070178de2
Merge pull request #433 from blindcrone/intercompatibility
|
9 сар өмнө |
Nel Nibcord
|
9712d696a9
Added a small script to compile grpc
|
9 сар өмнө |
Nel Nibcord
|
b787c676de
Updated unit tests
|
9 сар өмнө |
Daniel Newman
|
6d12deab2a
add better error handling:
|
9 сар өмнө |