 Aleksandr Borzunov
|
ddcda02b06
Hardcode IPs until DNS issues get resolved
|
2 anos atrás |
 Alexander Borzunov
|
b1ff8bdd6c
Bump version to 2.0.0.post1 (#384)
|
2 anos atrás |
|
Alexander Borzunov
|
e9a20e7e53
Require accelerate>=0.20.3 as transformers do (#383)
|
2 anos atrás |
|
Alexander Borzunov
|
057a2fb5de
Support Llama 2 (#379)
|
2 anos atrás |
|
Alexander Borzunov
|
3218534745
Fix --token arg (#378)
|
2 anos atrás |
 justheuristic
|
398a384075
Inherit bitsandbytes compute dtype correctly (override peft quirk) (#377)
|
2 anos atrás |
|
justheuristic
|
5a8de2f1f8
Fix handler memory leak, get rid of mp.Manager (#373)
|
2 anos atrás |
|
Alexander Borzunov
|
895327a0ae
Fix readme code example, require Python < 3.11 until supported (#374)
|
2 anos atrás |
|
Alexander Borzunov
|
c735dd7ba3
Update transformers to 4.31.0 and peft to 0.4.0 (#371)
|
2 anos atrás |
|
justheuristic
|
1ab35c2826
Typo in inference_session.py
|
2 anos atrás |
|
Alexander Borzunov
|
a6fdfc0556
Fix AssertionError on rebalancing (#370)
|
2 anos atrás |
|
Alexander Borzunov
|
f97582fb5f
Require transformers < 4.31.0 until we're compatible (#369)
|
2 anos atrás |
|
Alexander Borzunov
|
3b300c32e4
Update readme to show new models (#365)
|
2 anos atrás |
|
Alexander Borzunov
|
62d9ed5ce7
Implement shortest-path routing for inference (#362)
|
2 anos atrás |
 Ikko Eltociear Ashimine
|
fd30f7ce10
Fix typo in generation_algorithms.py (#364)
|
2 anos atrás |
|
Alexander Borzunov
|
11f0d992d7
Report inference, forward, and network RPS separately (#358)
|
2 anos atrás |
|
Alexander Borzunov
|
9517dd1e3d
Update readme and "Getting started" link (#360)
|
2 anos atrás |
|
Alexander Borzunov
|
3f733a96e3
Use bitsandbytes 0.40.1.post1 (#357)
|
2 anos atrás |
|
Alexander Borzunov
|
81c4a45ca2
Make a server ping next servers (#356)
|
2 anos atrás |
|
Alexander Borzunov
|
2c8959e713
Share more info about a server in DHT (#355)
|
2 anos atrás |
|
justheuristic
|
37fdcb3fe0
Switch adapters slightly faster (#353)
|
2 anos atrás |
|
Alexander Borzunov
|
9703358df0
Fix bugs in _choose_num_blocks() added in #346 (#354)
|
2 anos atrás |
|
Alexander Borzunov
|
1a78638c02
Test that bitsandbytes is not imported when it's not used (#351)
|
2 anos atrás |
|
justheuristic
|
c511990236
Remove unused import os (#352)
|
2 anos atrás |
|
Alexander Borzunov
|
e12d4c666b
Spam less in server logs (#350)
|
2 anos atrás |
|
justheuristic
|
010857a834
Estimate adapter memory overhead in choose_num_blocks() (#346)
|
2 anos atrás |
|
Alexander Borzunov
|
f605f093f7
Support LLaMA repos without "-hf" suffix (#349)
|
2 anos atrás |
|
Alexander Borzunov
|
90fbaab61e
Fix Docker build by avoiding Python 3.11 (#348)
|
2 anos atrás |
|
Alexander Borzunov
|
43acfe52a7
Import petals.utils.peft only when needed to avoid unnecessary import of bitsandbytes (#345)
|
2 anos atrás |
|
Alexander Borzunov
|
294970fe18
Update Colab link
|
2 anos atrás |
