bigscience-workshop petals issues

bigscience-workshop / petals

🌸 Run LLMs at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading

https://petals.dev

MIT License

8.89k stars 489 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Petals doesn't deal with server failure properly

#587 oldcpple opened 4 days ago
0
Donating System Memory?

#586 NavodPeiris opened 1 week ago
0
batch processing/parallel processing

#585 oldcpple opened 1 week ago
0
test

#584 xtinkt closed 2 weeks ago
0
Update transformers to 4.41.2

#583 xtinkt closed 1 week ago
0
Feature/kan 162 public ip

#582 jmikedupont2 closed 3 weeks ago
0
System_prompt

#581 EvilSumrak2049 opened 3 weeks ago
1
Setup and env.example

#580 Bakobiibizo closed 2 months ago
0
Feature/dht1

#579 jmikedupont2 closed 2 months ago
0
LLama-3-70B support

#578 ELigoP closed 2 months ago
3
Restrict PyTorch version to <2.3.0 to resolve import error

#577 Priyanshupareek closed 2 months ago
2
Error with PyTorch 2.3.0: Missing '_refresh_per_optimizer_state' in 'torch.cuda.amp.grad_scaler'

#576 Priyanshupareek closed 2 months ago
0
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument tensors in method wrapper_CUDA_cat)

#575 jmikedupont2 closed 2 months ago
2
Fix dummy cache allocation

#574 artek0chumak closed 2 months ago
0
Manual management of shards

#573 nrs-status opened 2 months ago
0
To test 1

#572 jmikedupont2 closed 2 months ago
0
DynamicCache and Beam Search

#571 artek0chumak opened 2 months ago
0
Fix Mixtral-related issues

#570 artek0chumak closed 2 months ago
0
Error trying to raise Mixtral private swarm server

#569 Qessia closed 2 months ago
13
Is there a way to shard a model without downloading it first?

#568 nrs-status closed 3 months ago
2
WIP GCP TPU XLA Build

#567 jmikedupont2 closed 2 months ago
0
compile to webassembly

#566 Donovoi opened 3 months ago
3
Latest bump "Bump transformers and accelerate versions (#554)" looks to destroy Falcon support.

#565 STEMBytes opened 3 months ago
5
Grok | Mixture-of-Experts | Model Support

#564 tibrezus opened 3 months ago
3
Fix p2p pushing in rpc_inference, support transformers 4.38.2

#563 justheuristic closed 3 months ago
0
Reachability issue while connecting to private swarm

#562 CherukupalleNaveen closed 3 months ago
1
content of 'labels' when doing prompt tuning of llama-2 on QA

#561 caochuxueeee opened 3 months ago
0
Direct server-to-server communication during finetuning

#560 dvmazur opened 3 months ago
0
Forward backward

#559 dvmazur closed 3 months ago
0
Clean disk space in push-docker-image.yaml

#558 justheuristic closed 4 months ago
0
Peft update

#557 artek0chumak opened 4 months ago
0
Feature Request: Distributed inference API where the "miners" are paid.

#556 NewtonTrendy opened 4 months ago
2
Fix push-docker-image's "No space left on device"

#555 dvmazur opened 4 months ago
0
Bump transformers and accelerate versions

#554 dvmazur closed 4 months ago
1
Add Mixtral models

#553 artek0chumak closed 3 months ago
0
Bump transformers

#552 dvmazur closed 5 months ago
0
Reachability Issue for private swarm

#551 VarunJoshi10 closed 4 months ago
1
Can not use direct server-to-server communication

#550 miaoqijun closed 3 months ago
6
Error when trying to launch private swarm using locally stored model

#549 sdryluth opened 6 months ago
0
Is there any plan to support MoE models like Mixtral8×7B?

#548 arronKler closed 2 months ago
5
Example of agents usage with Petals

#547 RomaA2000 opened 6 months ago
0
Add "Podman" usage to the documentation

#546 metal3d opened 6 months ago
0
replaced call to `_prepare_decoder_attention_mask()` with `_prepare_4d_causal_attention_mask()`

#545 poedator opened 7 months ago
1
install script to help the project deployment or for easy use

#544 elsolo5000-2 opened 7 months ago
0
Bump version for inference diagnostics

#543 justheuristic closed 7 months ago
0
ImportError: cannot import name 'AutoDistributedModelForCausalLM' from partially initialized module 'petals' (most likely due to a circular import)

#542 lacksfish closed 7 months ago
3
Prepull model data on private swarm

#541 wolfganghuse closed 7 months ago
5
M1 macOS installation error ("failed to build wheel for hivemind") with mac-native petals

#540 MinaAlmasi closed 7 months ago
2
Hotfix: require peft version 0.5.0

#539 justheuristic closed 7 months ago
0
Fix transformers version temporarily

#538 justheuristic closed 7 months ago
0