 Alexander Borzunov
|
8c546d988a
Test Llama, rebalancing, throughput eval, and all CLI scripts (#452)
|
há 2 anos atrás |
 justheuristic
|
5af04524dd
Split long sequences into chunks (#403)
|
há 2 anos atrás |
|
Alexander Borzunov
|
eb0664b993
Support Python 3.11 (#393)
|
há 2 anos atrás |
|
Alexander Borzunov
|
c735dd7ba3
Update transformers to 4.31.0 and peft to 0.4.0 (#371)
|
há 2 anos atrás |
 Artem Chumachenko
|
b9f0a5467f
Support peft LoRA adapters (#335)
|
há 2 anos atrás |
|
Alexander Borzunov
|
cb3f018f9f
Add LLaMA support (#323)
|
há 2 anos atrás |
 Max Ryabinin
|
5c0733711a
Use number of tokens for attn_cache_size (#286)
|
há 2 anos atrás |
|
Alexander Borzunov
|
74d8cda8c4
Add Python 3.10 to CI (#299)
|
há 2 anos atrás |
|
Alexander Borzunov
|
a7d3d02194
Fix invalid author email in setup.cfg (#287)
|
há 2 anos atrás |
|
justheuristic
|
c4938bc23e
Merge inference pools into one to increase inference speed (#225)
|
há 2 anos atrás |
|
Alexander Borzunov
|
702bb5a2c2
CI: Update deprecated actions, don't measure network RPS (#215)
|
há 2 anos atrás |
|
Alexander Borzunov
|
825f5dbf2d
CI: Convert model only when convert_model.py or setup.cfg change (#213)
|
há 2 anos atrás |
|
Alexander Borzunov
|
5ff250bee9
Improve errors in case of missing blocks, suggest to join your own server (#212)
|
há 2 anos atrás |
|
justheuristic
|
ae9e71fe8e
Add local tensor-parallel fwd/bwd (#143)
|
há 2 anos atrás |
|
Max Ryabinin
|
a0e8bbd28d
Fix arguments in remove_old_models.py (#153)
|
há 2 anos atrás |
|
justheuristic
|
a2066a4096
Optimize RemoteSequenceManager (#106)
|
há 2 anos atrás |
|
Alexander Borzunov
|
7bd5916744
Make Petals a pip-installable package (attempt 2) (#102)
|
há 2 anos atrás |
|
Alexander Borzunov
|
dc71574a63
Use public swarm by default (#92)
|
há 2 anos atrás |
|
justheuristic
|
e92487e5d2
Update dependency versions (#71)
|
há 2 anos atrás |
|
justheuristic
|
892d18fea7
Build cpuonly from bitsandbytes main (#70)
|
há 2 anos atrás |
|
justheuristic
|
d271b75dd4
Let users specify sequence length instead of assuming 2048 (#52)
|
há 3 anos atrás |
|
justheuristic
|
a2634001e9
Reduce vocabulary size in test model, fix bug in routing when overlapped (#45)
|
há 3 anos atrás |
|
justheuristic
|
9460220a10
make pytest outputs more verbose (#44)
|
há 3 anos atrás |
 Dmitry Baranchuk
|
c4aa1f49df
Rename 350m -> 560m (#43)
|
há 3 anos atrás |
|
Dmitry Baranchuk
|
11a424837f
integrate mixed-8bit model (#39)
|
há 3 anos atrás |
|
justheuristic
|
f0cffbf67e
Miscellaneous fixes to automatic tests (#35)
|
há 3 anos atrás |
|
justheuristic
|
f0c7383181
Implement RemoteSequential slicing and extra repr, add tests (#30)
|
há 3 anos atrás |
|
justheuristic
|
177d81bea6
CI: use GIT_REF_NAME instead of GIT_HEAD_REF (#28)
|
há 3 anos atrás |
|
justheuristic
|
49242d8100
[WIP] Fix CI runs in master (#27)
|
há 3 anos atrás |
|
justheuristic
|
01ed4db750
Fix default branch in CI (#26)
|
há 3 anos atrás |
