 Alex Cheema
|
54c98607ef
more robust grpc discovery with asyncio and proper error handling, add flops to device capabilities. fixes #23 and progress on #33
|
vor 1 Jahr |
|
Alex Cheema
|
fa9d416955
rm unused imports
|
vor 1 Jahr |
|
Alex Cheema
|
0af164f076
remove old PartitioningStrategy
|
vor 1 Jahr |
|
Alex Cheema
|
1b194b43f6
reference the code for each feature listed in README
|
vor 1 Jahr |
|
Alex Cheema
|
945f90f676
allow overriding inference_engine and separate flag for TINYGRAD_DEBUG
|
vor 1 Jahr |
|
Alex Cheema
|
47163d22db
broadcast results concurrently fixes #31
|
vor 1 Jahr |
 Alex Cheema
|
621a5f5d7b
Add license badge
|
vor 1 Jahr |
|
Alex Cheema
|
46d618abed
tiny fixes
|
vor 1 Jahr |
|
Alex Cheema
|
d4f55002ea
sort topology by memory descending (works well for now to workaround #12
|
vor 1 Jahr |
|
Alex Cheema
|
071b1caa0b
drop exo to 0.0.1 (still experimental)
|
vor 1 Jahr |
|
Alex Cheema
|
e7dcdac24c
fix exo text
|
vor 1 Jahr |
|
Alex Cheema
|
17ecf26677
typo bullet point
|
vor 1 Jahr |
|
Alex Cheema
|
9958ac3917
Make known issues more prominent
|
vor 1 Jahr |
|
Alex Cheema
|
72fe293729
exo text on start and stop
|
vor 1 Jahr |
|
Alex Cheema
|
fbbb45c37e
install script
|
vor 1 Jahr |
|
Alex Cheema
|
3778301b47
add alternative installation through install.sh
|
vor 1 Jahr |
|
Alex Cheema
|
4f4696e09d
remove calls to updateTotalTokens in tiny, not sure why its there
|
vor 1 Jahr |
|
Alex Cheema
|
d4e0a7d14b
add endpoint to get number of encoded tokens
|
vor 1 Jahr |
|
Alex Cheema
|
127b8e011c
dont explicitly specify show_index
|
vor 1 Jahr |
|
Alex Cheema
|
a94bdbb9bd
serve tinychat static
|
vor 1 Jahr |
|
Alex Cheema
|
e82dab1d4c
remove tinygrad hidden files
|
vor 1 Jahr |
|
Alex Cheema
|
c2fcee4387
only retail examples/tinychat from tinygrad subtree
|
vor 1 Jahr |
|
Alex Cheema
|
e1cb840b31
Merge commit '0870e6bfb0d46d58e29c6d5822e8bf629d03bdcc' as 'tinychat'
|
vor 1 Jahr |
|
Alex Cheema
|
0870e6bfb0
Squashed 'tinychat/' content from commit fa7e734b4
|
vor 1 Jahr |
|
Alex Cheema
|
d8c40bb4f1
print a warning if stream task ever times out
|
vor 1 Jahr |
|
Alex Cheema
|
1e1e11cdc6
check if inference_engine has tokenizer before printing with it
|
vor 1 Jahr |
|
Alex Cheema
|
8df2f4d857
support for streaming from non-tail nodes on chatgpt api, addresses #20
|
vor 1 Jahr |
|
Alex Cheema
|
8762effaf4
chatgpt api repsonse streaming solves #20
|
vor 1 Jahr |
|
Alex Cheema
|
5de2ea51f5
default to llama-3-8b and temperature=0 if not provided
|
vor 1 Jahr |
|
Alex Cheema
|
5c3f0e3ac6
faster initial node discovery
|
vor 1 Jahr |
