DEDUPED Hot AI repos, turned into use cases Get 5 a day

What you can do with Halo

Halo is an open-source framework from White Circle for training large language and multimodal models with pre-training, SFT, preference optimization, and multi-turn RL on Hugging Face models.

Trended earlier

whitecircle/halo on GitHub

Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk.

What do you want to do?

5 use cases

  • Fine-tune a Hugging Face LLM with LoRA on one GPU

    Run a LoRA SFT job on a Qwen3 model with one command from a ready YAML config.

    View prompt
    In the whitecircle/halo repo, run `halo launch sft <target-config>` inside the Halo Docker image on one GPU, using examples/sft/qwen3/qwen3-4b-ultrachat-lora.yaml as the default. Expected output: a LoRA checkpoint written by the run. Verify first with `--dry-run` that the printed command matches the config path you chose.

  • Train a mixture-of-experts model across 8 GPUs

    Use Expert Parallelism on a MoE model to spread experts over 8 GPUs with one launch command.

    View prompt
    In the whitecircle/halo repo, launch the MoE SFT example with `halo launch sft <target-config> -n 8 --expert_parallel_size=8` inside the Halo image on an 8-GPU node. Expected output: a completed run with expert shards in the output directory. Verify with `--dry-run` that the EP setting appears in the printed command before the real run.

  • Run multi-turn RL against vLLM or SGLang rollouts

    Train a policy with asynchronous GRPO where Ray actors roll out environments on vLLM or SGLang.

    View prompt
    In the whitecircle/halo repo, read human-docs/training-methods/async-grpo-environments.md and then launch the environmental-grpo method with <target-config> inside the Halo image. Expected output: a running async GRPO job that reports rollouts and weight syncs. Verify with `halo launch --list` that environmental-grpo is available before launching.

  • Quantize a trained checkpoint to NVFP4

    Convert a saved checkpoint to a low-precision NVFP4 format with one Halo run command.

    View prompt
    In the whitecircle/halo repo, run `halo run quantize-to-lowp` with --input_dir <checkpoint_dir> and --output_dir <quantized_dir> and --format nvfp4 inside the Halo image. Expected output: a quantized checkpoint in the output directory. Verify that the input and output paths are different directories before running.

  • Let a coding agent set up and launch training jobs

    Point Claude Code or Codex at the repo skills to choose parallelism and launch a Halo run.

    View prompt
    In the whitecircle/halo repo, use the skills/parallelism and skills/launch guides to choose settings for <target-config> and then run `halo launch sft` with --dry-run inside the Halo image. Expected output: a printed launch command with the chosen parallelism flags. Verify that the flags match one of the allowed EP, CP, TP, or ETP combinations before any real run.

How to use Halo

  1. Check you have what it needs: Docker installed on a host with an NVIDIA GPU (Hopper H100/H200 or Blackwell B200/B300 images are published); Network access to public.ecr.aws to pull the Halo image anonymously; A Hugging Face model or the example configs in the examples/ folder.
  2. Open whitecircle/halo on GitHub and read the README and license.
  3. Pull the image that matches your GPU: docker pull public.ecr.aws/whitecircle/halo:hopper for H100/H200 or docker pull public.ecr.aws/whitecircle/halo:blackwell for B200/B300.
  4. Start a container from that image with GPU access and open a shell in the repo. (inferred, check the README)
  5. Print the command without running it: halo launch sft examples/sft/qwen3/qwen3-4b-ultrachat-lora.yaml --dry-run. (inferred, check the README)
  6. Run the job: halo launch sft examples/sft/qwen3/qwen3-4b-ultrachat-lora.yaml.

These steps come from our top use case, Fine-tune a Hugging Face LLM with LoRA on one GPU, which also has a copy-paste agent prompt.

Halo alternatives · Best AI repos for Software and DevOps

About this repo

  • Who it is forWritten for ML engineers and AI agents who have access to NVIDIA Hopper or Blackwell GPUs and want to train or fine-tune Hugging Face models at scale.
  • Why it is hot
  • LanguagePython
  • Sourcegithub.com/whitecircle/halo · Back to Hot repos

Requirements

  • Docker installed on a host with an NVIDIA GPU (Hopper H100/H200 or Blackwell B200/B300 images are published)
  • Network access to public.ecr.aws to pull the Halo image anonymously
  • A Hugging Face model or the example configs in the examples/ folder
  • Enough GPU memory for the chosen model and method (the README shows single-GPU LoRA and 8-GPU runs)
Maintain Halo? Add the Featured on Deduped badge

Paste this into your README:

[![Featured on Deduped](https://dedupedai.com/badge/whitecircle-halo.svg)](https://dedupedai.com/use-cases/whitecircle-halo/?ref=badge)

More badge styles and HTML

Questions about Halo

What is Halo?

Halo is an open-source framework from White Circle for training large language and multimodal models with pre-training, SFT, preference optimization, and multi-turn RL on Hugging Face models. It is an open-source project on GitHub: https://github.com/whitecircle/halo.

Who is Halo for?

Written for ML engineers and AI agents who have access to NVIDIA Hopper or Blackwell GPUs and want to train or fine-tune Hugging Face models at scale.

What can I do with Halo?

5 use cases so far, for example: Fine-tune a Hugging Face LLM with LoRA on one GPU; Train a mixture-of-experts model across 8 GPUs; Run multi-turn RL against vLLM or SGLang rollouts. Each has steps and a copy-paste agent prompt.

What do I need to run Halo?

Docker installed on a host with an NVIDIA GPU (Hopper H100/H200 or Blackwell B200/B300 images are published); Network access to public.ecr.aws to pull the Halo image anonymously; A Hugging Face model or the example configs in the examples/ folder; Enough GPU memory for the chosen model and method (the README shows single-GPU LoRA and 8-GPU runs).

What are alternatives to Halo?

Other hot AI repos in Software and DevOps: Harness2, iFixAi, Orca, Diagram Design, AI Memory.

Is Halo safe to run?

Deduped does not own, audit or vouch for it. Read the README and license, run it on your own systems first and review every command before you run it.