bigscience-workshop/petals

Issues

• AttributeError: module 'numpy' has no attribute 'ndarray'

  #518 opened a year ago
  5

• Unable to acquire fcntl.LOCK_SH lock when cache directory is mounted on NFS

  #515 opened a year ago
  0

• Can petals run a quantized (use GPTQ) llama model ?

  #514 opened a year ago
  3

• Add YaLM-100B

  #512 opened a year ago
  1

• [might be bug?] Failed to connect to bootstrap peers when using docker image on truenas scale

  #511 opened a year ago
  4

• Petals server Network ports remain open after python process shutdown

  #509 opened a year ago
  1

• No target modules when attempting LoRA (peft) training

  #508 opened a year ago
  1

• Extend ServerInfo with (start_block, end_block)

  #507 opened a year ago
  1

• Issue with beam search decoding

  #503 opened a year ago
  2

• More powerful session API

  #495 opened a year ago
  1

• Incentive system based on Lightning

  #494 opened a year ago
  1

• Need help with text generation adapter finetune

  #492 opened a year ago
  0

• Can't install on windows

  #488 opened a year ago
  9

• How can we make this work long term?

  #483 opened a year ago
  2

• `model.generate(input_ids=...)` support

  #481 opened a year ago
  1

• Python the only option to run?

  #469 opened a year ago
  2

• Random error when starting docker container again

  #468 opened a year ago
  0

• how to avoid this server failure? Seems to happen randomly after 1 hour of running a script.

  #466 opened a year ago
  1

• Out of memory on a client with 8 GB RAM

  #465 opened a year ago
  4

• Add repetition_penalty to RemoteGenerationMixin

  #460 opened a year ago
  3

• Add server option to prioritize a list of given peer IDs (so you can prioritize your own clients)

  #457 opened 2 years ago
  0

• Hope to get in touch

  #455 opened 2 years ago
  0

• OS-level lock for choosing blocks

  #453 opened 2 years ago
  0

• Repack StableBeluga2 to small shards + safetensors

  #450 opened a year ago
  6

• Make client-side macOS support

  #449 opened a year ago
  1

• Report average queue size in tokens (per last 10 min) for routing

  #447 opened 2 years ago
  0

• API reference

  #446 opened 2 years ago
  0

• Increase file limit automatically

  #444 opened a year ago
  1

• GGML + petals

  #439 opened 2 years ago
  1

• local time must be within 3 seconds of others

  #433 opened 2 years ago
  2

• RuntimeError: "addmm_impl_cpu_" not implemented for 'Half'

  #432 opened 2 years ago
  1

• PingAggregator returns inf for all servers that use relays

  #429 opened 2 years ago
  1

• How to parallelize this code for model.generate?

  #427 opened 2 years ago
  0

• Ask for help and discuss Petals in our Discord!

  #426 opened 2 years ago
  0

• Add an example on how decoding algorithm can be used.

  #425 opened a year ago
  5

• Problem running petals on virtual CPU

  #421 opened 2 years ago
  4

• Request to Enable Git Discussions Feature

  #420 opened 2 years ago
  3

• Conversational Agent with Llama-2-70b-chat-hf

  #419 opened 2 years ago
  4

• eidt. may already exist in server.py Next Step create a distributed hash table to indicate which IP addresses are hosting which parts of different models rather than using a redis database

  #417 opened 2 years ago
  2

• how to get only embeddings?

  #416 opened 2 years ago
  1

• Remember Chat History

  #415 opened 2 years ago
  1

• How to get faster inference?

  #414 opened 2 years ago
  1

• CUDA out of memory (Petals Server via Docker)

  #413 opened 2 years ago
  2

• [feature] add privacy

  #412 opened 2 years ago
  1

• Expand on FLUX network of decentralized nodes

  #408 opened 2 years ago
  0

• RuntimeError: Attempt to start a new process before current process has finished its bootstrapping phase

  #405 opened a year ago
  6

• ERROR: Failed building wheel for grpcio-tools Running setup.py clean for grpcio-tools Failed to build grpcio-tools ERROR: Could not build wheels for grpcio-tools, which is required to install pyproject.toml-based projects

  #401 opened 2 years ago
  2

• Add windows support

  #400 opened a year ago
  7

• Petals Error: GPU is not available

  #398 opened 2 years ago
  3

• Cuda 11.8 supported?

  #395 opened 2 years ago
  2