 Alex Cheema
|
c4238e7d25
Merge pull request #151 from sammcj/patch-2
|
il y a 1 an |
 Alex Cheema
|
9b8e1bcddc
trigger test
|
il y a 1 an |
 Sam
|
1819df36f5
Add common RTX A series cards to device_capabilities.py
|
il y a 1 an |
|
Alex Cheema
|
a930be4fd9
t
|
il y a 1 an |
|
Alex Cheema
|
53ec180d40
fix test import
|
il y a 1 an |
|
Alex Cheema
|
611085b38d
trigger test
|
il y a 1 an |
|
Alex Cheema
|
2a214db7a4
rm tokenizer from test
|
il y a 1 an |
|
Alex Cheema
|
803dffd1c4
always call convert_from_huggingface with tinygrad models. this was broken by shard layer filtering which made the check sometimes fail. fixes #144
|
il y a 1 an |
|
Alex Cheema
|
a8b877bcd8
Update README.md
|
il y a 1 an |
|
Alex Cheema
|
7ddb80e245
f-string expression part cannot include a backslash fixes #142
|
il y a 1 an |
|
Alex Cheema
|
781a71ccff
Merge pull request #138 from exo-explore/better_downloads
|
il y a 1 an |
|
Alex Cheema
|
75681a9707
use async for all file ops, cache fetch_file_list, cache commit hash, quickly check file sizes on disk before making requests
|
il y a 1 an |
|
Alex Cheema
|
6c1bf127b3
add --max-parallel-downloads flag that limits the number of downloads at a time with asyncio.semaphore
|
il y a 1 an |
|
Alex Cheema
|
440fd35ea7
upgrade aiohttp
|
il y a 1 an |
|
Alex Cheema
|
8e6414b257
spacing
|
il y a 1 an |
|
Alex Cheema
|
e8267e7387
LAPTOP GPU and Laptop GPU prefixes
|
il y a 1 an |
|
Alex Cheema
|
31641d1023
tinygrad select model size
|
il y a 1 an |
|
Alex Cheema
|
7b704e8d98
Merge pull request #136 from exo-explore/exo_interfaces
|
il y a 1 an |
|
Alex Cheema
|
e6902b2fcf
add --download-quick-check flag to bypass the hf api calls / remote file checks
|
il y a 1 an |
|
Alex Cheema
|
84afdbcbe8
add --download-quick-check flag to bypass the hf api calls / remote file checks
|
il y a 1 an |
|
Alex Cheema
|
1f736dacd7
Merge pull request #135 from exo-explore/exo_interfaces
|
il y a 1 an |
|
Alex Cheema
|
71591d2ebc
display all interfaces web chat and chatgpt api are available on fixes #134
|
il y a 1 an |
|
Alex Cheema
|
3bd5a116df
ignore files that dont match allow patterns
|
il y a 1 an |
|
Alex Cheema
|
3db3e8294c
make download panel slightly larger
|
il y a 1 an |
|
Alex Cheema
|
9e78c42b4b
Merge pull request #124 from exo-explore/refactor_model_download
|
il y a 1 an |
|
Alex Cheema
|
5112f53a37
trigger ci
|
il y a 1 an |
|
Alex Cheema
|
047ef48c1d
use separate hf cache dirs for chatgpt api integration test. its an unusual setup where we're running 2 exo instances on the same device which share a disk and hf cache
|
il y a 1 an |
|
Alex Cheema
|
2be446546f
refactor tinygrad, only load necessary layers for each shard fixes #128, enable JIT (much faster), prefill all layers not just the first shard fixes #12, use new ShardDownloader for more robust, parallel downloads
|
il y a 1 an |
|
Alex Cheema
|
357331c55f
remove some logs, make get_allow_patterns out of class
|
il y a 1 an |
|
Alex Cheema
|
b1eb05ed47
debug level 7 for tests
|
il y a 1 an |
