Building Moon Bot: A Slack-Native Coding Agent Backed by HuggingFace Buckets
⢠53
AI & ML interests
The AI community building the future.
Recent Activity
View all activity
Papers
Can We Defend Against AI-Generated Video Attacks on Real-World Crisis Events? A Systematic Evaluation of Detectors, Generators and Social Dissemination
Seeing the Needle in the Haystack: Towards Weakly-Supervised Log Instance Anomaly Localization via Counterfactual Perturbation
Articles
lysandreÂ
updated a
dataset about 4 hours ago
nielsrÂ
updated a
bucket about 8 hours ago
tarekziadeÂ
updated a
bucket about 11 hours ago
alvarobarttÂ
updated a
dataset about 12 hours ago
sayakpaulÂ
authored a
paper 12 days ago
sayakpaulÂ
submitted a
paper to Daily Papers 12 days ago
Post
1980
Hugging Face MCP Server v0.4.9
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Attach images from hf_fs tool.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Attach images from hf_fs tool.
Post
5233
who's working on an NVFP4 version of Kimi-K3?
- 4 replies
Post
2416
432 GB of ultra-fast HBM4 and up to 23.3 TB/s of memory bandwidth on a single GPU đ¤Ż.
Two weeks ago, we got early access to AMD's new Instinct MI455X, and our first goal was simple: make sure đ¤ Transformers works on day one.
Over the past few weeks, we worked closely with the AMD team to validate the platform, enable Flash Attention, add torchcodec support for multimodal models, and resolve issues uncovered during testing.
The result:
â 99.5% success rate across our 24 core Transformers model architectures - already on par with our daily CI on previous AMD and NVIDIA platforms.
The hardware is just as exciting. With 432 GB of HBM per GPU, our early capacity experiments showed more than 3Ă the concurrent long-context requests compared to MI300, thanks to the much larger KV cache capacity.
A huge thanks to the AMD team for the early access and the great collaboration!
Read the full blog đ
https://huggingface.co/blog/badaoui/transformers-on-amd-mi455
Two weeks ago, we got early access to AMD's new Instinct MI455X, and our first goal was simple: make sure đ¤ Transformers works on day one.
Over the past few weeks, we worked closely with the AMD team to validate the platform, enable Flash Attention, add torchcodec support for multimodal models, and resolve issues uncovered during testing.
The result:
â 99.5% success rate across our 24 core Transformers model architectures - already on par with our daily CI on previous AMD and NVIDIA platforms.
The hardware is just as exciting. With 432 GB of HBM per GPU, our early capacity experiments showed more than 3Ă the concurrent long-context requests compared to MI300, thanks to the much larger KV cache capacity.
A huge thanks to the AMD team for the early access and the great collaboration!
Read the full blog đ
https://huggingface.co/blog/badaoui/transformers-on-amd-mi455
- 1 reply
sayakpaulÂ
authored a
paper about 2 months ago
Post
1406
Hugging Face MCP Server v0.3.29
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Included "papers" in the new hf_fs tool. Includes listing of trending/daily.
This is a new tool under observation - disable the "Paper Semantic Search" tool for best results.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Included "papers" in the new hf_fs tool. Includes listing of trending/daily.
This is a new tool under observation - disable the "Paper Semantic Search" tool for best results.
hf://papers/
âââ README.md
âââ daily/
â âââ latest
â âââ YYYY/
â âââ MM/
â âââ DD/
âââ trending/
âââ ARXIV_ID/
âââ metadata.json
âââ paper.md
âââ models/
âââ datasets/
âââ spaces/sayakpaulÂ
authored 3
papers 2 months ago
sayakpaulÂ
submitted a
paper to Daily Papers 2 months ago
albertvillanovaÂ
posted an update 2 months ago
Post
4010
đ KTO is now part of the stable TRL API
As of Promote KTO to stable API, KTOTrainer and KTOConfig have graduated from trl.experimental to the stable trl API. https://github.com/huggingface/trl/pull/6175
This one closes out a long road. Over the past 6+ months, the "Align KTO with DPO" effort landed ~90 PRs methodically bringing KTO up to the standard we hold for stable trainers, one carefully-scoped change at a time:
- Feature parity with DPO: full VLM support (incl. multi-image), sync_ref_model, PEFT + Liger, ZeRO-3 + PEFT dtype fix, pad_to_multiple_of, activation offloading, IterableDataset and dict eval_dataset, remove_unused_columns, and reference-logprob precomputation at init.
- Consistency with DPO: aligned method order and signatures, tokenization, _prepare_dataset, PEFT handling, ref-model preparation for distributed training, and config layout â plus a new DataCollatorForKTO and output format. Metrics moved into _compute_loss and simplified to direct averages via the shared _metrics attribute.
- Removing legacy baggage: dropped encoder-decoder support, BOS/EOS handling, null_ref_context, generate_during_eval, model_init, preprocess_logits_for_metrics, model/ref adapter names, and several dead config knobs.
- Coverage: a full test suite mirroring DPO, text collator tests, VLM tests, and slow tests.
- The promotion itself: the experimental â stable move (#6175) and shim cleanup (#6287), handled so downstream users get a clean deprecation path.
Honestly, this has been one of the more complex tasks I've taken on since joining the team, not because any single change was hard, but because it demanded sustained consistency across a ~2,000-line trainer, with every branch, comment, and edge case kept in lockstep with DPO.
Huge thanks to everyone who reviewed along the way (especially @qgallouedec ), the incremental review cadence is exactly what kept this maintainable.
KTO now sits on equal footing with our other flagship trainers. đ
As of Promote KTO to stable API, KTOTrainer and KTOConfig have graduated from trl.experimental to the stable trl API. https://github.com/huggingface/trl/pull/6175
This one closes out a long road. Over the past 6+ months, the "Align KTO with DPO" effort landed ~90 PRs methodically bringing KTO up to the standard we hold for stable trainers, one carefully-scoped change at a time:
- Feature parity with DPO: full VLM support (incl. multi-image), sync_ref_model, PEFT + Liger, ZeRO-3 + PEFT dtype fix, pad_to_multiple_of, activation offloading, IterableDataset and dict eval_dataset, remove_unused_columns, and reference-logprob precomputation at init.
- Consistency with DPO: aligned method order and signatures, tokenization, _prepare_dataset, PEFT handling, ref-model preparation for distributed training, and config layout â plus a new DataCollatorForKTO and output format. Metrics moved into _compute_loss and simplified to direct averages via the shared _metrics attribute.
- Removing legacy baggage: dropped encoder-decoder support, BOS/EOS handling, null_ref_context, generate_during_eval, model_init, preprocess_logits_for_metrics, model/ref adapter names, and several dead config knobs.
- Coverage: a full test suite mirroring DPO, text collator tests, VLM tests, and slow tests.
- The promotion itself: the experimental â stable move (#6175) and shim cleanup (#6287), handled so downstream users get a clean deprecation path.
Honestly, this has been one of the more complex tasks I've taken on since joining the team, not because any single change was hard, but because it demanded sustained consistency across a ~2,000-line trainer, with every branch, comment, and edge case kept in lockstep with DPO.
Huge thanks to everyone who reviewed along the way (especially @qgallouedec ), the incremental review cadence is exactly what kept this maintainable.
KTO now sits on equal footing with our other flagship trainers. đ
- 2 replies
jeffboudierÂ
posted an update 2 months ago
Post
445
@alvarobartt published a step-by-step guide to deploy zai-org/GLM-5.2 on AMD GPUs, using newly released features in Microsoft Foundry. The FP8 model fits on a single node of MI300X GPUs, which cuts the bill in half vs. H100.
https://alvarobartt.com/goal-glm-5.2-on-foundry/
https://alvarobartt.com/goal-glm-5.2-on-foundry/
Post
229
We have recently added Torch Stable ABI support to kernels and kernel-builder. This allows kernel developers to target a particular Torch version and the kernel will be supported on that Torch version and later Torch versions (up to ~2 years).
This makes it much easier to write kernels with long-term support and not just the last two Torch releases.
We have also started rolling out Stable ABI support to kernels in kernels-community, starting with Flash Attention 3, supporting Torch 2.9 and later as well as CUDA versions starting at 12.6:
https://huggingface.co/kernels/kernels-community/flash-attn3/tree/v1/build
This makes it much easier to write kernels with long-term support and not just the last two Torch releases.
We have also started rolling out Stable ABI support to kernels in kernels-community, starting with Flash Attention 3, supporting Torch 2.9 and later as well as CUDA versions starting at 12.6:
https://huggingface.co/kernels/kernels-community/flash-attn3/tree/v1/build
- 1 reply
craffelÂ
authored a
paper 3 months ago
cfahlgren1Â
submitted a
paper to Daily Papers 3 months ago