 Alex Cheema
|
6662d5668c
load mlx model shard on mlx thread so it doesnt block
|
7 mesi fa |
|
Alex Cheema
|
7c649085a1
fix eta/speed for resuming an existing download, using the session downloaded bytes
|
7 mesi fa |
|
Alex Cheema
|
90e0e2761f
ignore not_started progress updates
|
7 mesi fa |
|
Alex Cheema
|
265586f7b4
set timeout on get too
|
7 mesi fa |
|
Alex Cheema
|
4748bb7dc7
increase file download timeout to 30min
|
7 mesi fa |
|
Alex Cheema
|
ae770db4f3
increase download chunks to 1MB
|
7 mesi fa |
|
Alex Cheema
|
82f75d0ccf
increase hf download http timeout 15 mins for large downloads
|
7 mesi fa |
|
Alex Cheema
|
295f41c5cc
increase bench job timeout to give enough time to download
|
7 mesi fa |
|
Alex Cheema
|
19a27c5bfd
HF_HOME -> EXO_HOME
|
7 mesi fa |
|
Alex Cheema
|
d7ca9b7732
show each node id in the tinychat topology viz
|
7 mesi fa |
|
Alex Cheema
|
b349e48b0d
fix visual bug where frontend would show the full hf repo size, but in some cases that includes redundant files so we should use the model index in those cases too
|
7 mesi fa |
|
Alex Cheema
|
21586063f6
use llama-3.2-1b in tinygrad test
|
7 mesi fa |
|
Alex Cheema
|
277d63d860
special case when a model doesnt have a model index file, then use wildcard for allow_patterns
|
7 mesi fa |
|
Alex Cheema
|
74379ef671
log download logs with DEBUG>=6 very verbose
|
7 mesi fa |
|
Alex Cheema
|
3c7bd48aa3
get rid of some more hf bloat
|
7 mesi fa |
|
Alex Cheema
|
1df023023e
remove a lot of hf bloat
|
7 mesi fa |
|
Alex Cheema
|
b89495f444
rewrite ShardDownloader, simplify significantly
|
7 mesi fa |
|
Alex Cheema
|
a3766f538a
add exception for mlx-community/DeepSeek-R1-3bit and mlx-community/DeepSeek-V3-3bit in tokenizers test
|
7 mesi fa |
|
Alex Cheema
|
82ef086010
add deepseek-v3-3bit and deepseek-r1-3bit
|
7 mesi fa |
|
Alex Cheema
|
55ea366932
fix post_init deepseek v3
|
7 mesi fa |
 Alex Cheema
|
63318983de
Merge pull request #631 from sigseg5/main
|
7 mesi fa |
 sigseg5
|
fb841a1f50
Adjust truncate size in history list for text without any spaces
|
7 mesi fa |
|
sigseg5
|
4512366580
Fix bubble behavior when user passes long text without any spaces
|
7 mesi fa |
|
sigseg5
|
9525c0e7a7
Add adaptive padding for user and assistant messages on width <= 1480px
|
7 mesi fa |
|
Alex Cheema
|
66f73768cc
Merge pull request #627 from exo-explore/deepseek
|
7 mesi fa |
|
Alex Cheema
|
fdd05baddb
fix tokenizer tests
|
7 mesi fa |
|
Alex Cheema
|
59174bdc62
we have a lot of models so group them nicely
|
7 mesi fa |
|
Alex Cheema
|
cfdaaef8e6
handle thinking outputs nicely, format latex beautifully
|
7 mesi fa |
|
Alex Cheema
|
d8ffa59dba
add deepseek v1, v3 and all the distills
|
7 mesi fa |
|
Alex Cheema
|
aa1ce21f82
Merge pull request #625 from eltociear/patch-1
|
7 mesi fa |
