microsoft / microsoft/BitNet

pre-tuned kernel not found for model microsoft/BitNet-b1.58-2B-4T

Open
#230 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
40.3k
Forks
3.7k
PR merge metrics
No merged PRs in 30d

Description

Using the latest version, I'm trying to setup BitNet following the setup_env.py help instructions:

python .\setup_env.py -help
usage: setup_env.py [-h]
                    [--hf-repo {1bitLLM/bitnet_b1_58-large,1bitLLM/bitnet_b1_58-3B,HF1BitLLM/Llama3-8B-1.58-100B-tokens,tiiuae/Falcon3-7B-Instruct-1.58bit,tiiuae/Falcon3-7B-1.58bit,tiiuae/Falcon3-10B-Instruct-1.58bit,tiiuae/Falcon3-10B-1.58bit,tiiuae/Falcon3-3B-Instruct-1.58bit,tiiuae/Falcon3-3B-1.58bit,tiiuae/Falcon3-1B-Instruct-1.58bit,microsoft/BitNet-b1.58-2B-4T}]
                    [--model-dir MODEL_DIR] [--log-dir LOG_DIR] [--quant-type {i2_s,tl2}] [--quant-embd] [--use-pretuned]

Running the following command ran into an error however:

python setup_env.py --hf-repo microsoft/BitNet-b1.58-2B-4T -q i2_s --use-pretuned
ERROR:root:Pretuned kernels not found for model microsoft/BitNet-b1.58-2B-4T

It seems that the microsoft/ was not removed, and so didn't match with "BitNet-b1.58-2B-4T"

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start in setup_env.py at the --use-pretuned handling and trace how the --hf-repo value is matched to available pre-tuned kernels. Reproduce the command from the report and verify that microsoft/BitNet-b1.58-2B-4T is recognized instead of producing the missing-kernel error.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai-infra-agents
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
52/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.