 Alex Cheema
|
ea3322dea4
remove comment
|
11 月之前 |
|
Alex Cheema
|
e0fda94d20
use sets for shard specific patterns
|
11 月之前 |
|
Alex Cheema
|
b239c8a6d0
Merge branch 'main' into non_blocking
|
11 月之前 |
|
Alex Cheema
|
8f65e1e697
fix weight_map resolution. previously we were always defaulting to allow pattern *.safetensors
|
11 月之前 |
|
Alex Cheema
|
6881722b72
simplify non-blocking mlx inference
|
11 月之前 |
|
Alex Cheema
|
9db16f8dca
use a queue for non-blocking mlx inference
|
11 月之前 |
|
Alex Cheema
|
0ca5c26094
run mlx inference engine on a single thread too
|
11 月之前 |
|
Alex Cheema
|
58f535d0b0
formatting
|
11 月之前 |
|
Alex Cheema
|
2950373d36
experiment with tinygrad on its own thread, so it doesnt block event loop
|
11 月之前 |
|
Alex Cheema
|
41f0a22e76
DEBUG>=8 for SendOpaqueStatus logs
|
11 月之前 |
|
Alex Cheema
|
01cc6a4c9d
fix Mistral-Large special case when we pass in a path
|
11 月之前 |
|
Alex Cheema
|
41dd700ff4
less aggressive logs for opaque status / download progress. too much spam
|
11 月之前 |
 Alex Cheema
|
35aba75be6
Merge pull request #194 from exo-explore/better_networking
|
11 月之前 |
|
Alex Cheema
|
4537d614aa
circleci use tee to output logs in realtime as well as capture them
|
11 月之前 |
|
Alex Cheema
|
56c1bf9a95
consistent remove _secs / -secs suffix
|
11 月之前 |
|
Alex Cheema
|
f342cdcae9
get rid of -secs suffix
|
11 月之前 |
|
Alex Cheema
|
a0d9c90e96
shorten cli name --chatgpt-api-response-timeout
|
11 月之前 |
|
Alex Cheema
|
8cb678e795
better logs around peer connecting / disconnecting
|
11 月之前 |
|
Alex Cheema
|
c97da5480f
add id to set
|
11 月之前 |
|
Alex Cheema
|
80c48b9e76
update visited with self.id, timeout on collecting topology from a peer 5s
|
11 月之前 |
|
Alex Cheema
|
355c579965
more robust discovery / peer handling. now we track if the same node id changes address, then we immediately conenct to it
|
11 月之前 |
|
Alex Cheema
|
8114a79ede
add back listen and cleanup tasks
|
11 月之前 |
|
Alex Cheema
|
dcb3ac76a8
test kill pids
|
11 月之前 |
|
Alex Cheema
|
3dd81a1e05
fix UDPDiscovery params, create a new transport every time we broadcast
|
11 月之前 |
|
Alex Cheema
|
15b5043d6e
test for reconnect
|
11 月之前 |
|
Alex Cheema
|
baf6efd321
cleaner discovery
|
11 月之前 |
|
Alex Cheema
|
8baaad7f6b
Merge branch 'main' into better_networking
|
11 月之前 |
|
Alex Cheema
|
57215041a0
todo for speculative model
|
11 月之前 |
|
Alex Cheema
|
dc3b2bde39
use NousResearch/Meta-Llama-3.1-70B-Instruct as tinygrad llama-3.1-70b model, previously using non-instruct model
|
11 月之前 |
|
Alex Cheema
|
12609cb6e4
integration test for udp discovery with grpc server
|
11 月之前 |
