Submissions from huggingface.co

		DeepSeek V4 Flash (huggingface.co)
		13 points by S0y 8 hours ago \| past \| discuss
		DeepSeek-V4 (huggingface.co)
		7 points by meetpateltech 9 hours ago \| past \| discuss
		[dupe] DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence (huggingface.co)
		152 points by cmrdporcupine 9 hours ago \| past \| 15 comments
		DeepSeek-V4 Technical Report [pdf] (huggingface.co)
		24 points by tianyicui 9 hours ago \| past \| 4 comments
		OpenAI Privacy Filter (huggingface.co)
		4 points by handfuloflight 1 day ago \| past \| discuss
		The PR you would have opened yourself (huggingface.co)
		1 point by gmays 1 day ago \| past \| discuss
		Qwen/Qwen3.6-27B · Hugging Face (huggingface.co)
		40 points by cgeier 1 day ago \| past \| 6 comments
		Show HN: Image to Explorable 3D Scene (huggingface.co)
		2 points by ggamecrazy 2 days ago \| past \| discuss
		Building a Fast Multilingual OCR Model with Synthetic Data (huggingface.co)
		3 points by gmays 2 days ago \| past \| discuss
		We OCR'ed 30k papers using Codex, open OCR models and Jobs (huggingface.co)
		7 points by speckx 2 days ago \| past \| discuss
		5.6x throughput on Kimi K2.6 by speculating less (huggingface.co)
		11 points by florianleibert 3 days ago \| past \| 2 comments
		Granite-4.0-Tiny-Preview (huggingface.co)
		1 point by handfuloflight 3 days ago \| past \| discuss
		Kimi K2.6 (huggingface.co)
		11 points by kbumsik 3 days ago \| past \| 1 comment
		Show HN: We trained a 32B model to beat Opus 4 at credit card optimization (huggingface.co)
		6 points by dimberman 3 days ago \| past \| discuss
		Show HN: Overlapping Speaker Transcription Model (huggingface.co)
		1 point by mcgov 4 days ago \| past \| discuss
		Qwen releases Qwen3-Embedding-0.6B (huggingface.co)
		4 points by arabicalories 4 days ago \| past \| 2 comments
		Why do language models feel worse even as benchmarks improve? [pdf] (huggingface.co)
		2 points by scaledsystems 4 days ago \| past \| discuss
		How to Fine-Tune a Reasoning Model? (huggingface.co)
		1 point by Anon84 5 days ago \| past \| discuss
		Parsebench open leaderboard for PDF parsing (huggingface.co)
		1 point by pierre 6 days ago \| past \| discuss
		Vakra: Reasoning, Tool Use, and Failure Modes of Agents (huggingface.co)
		2 points by gmays 7 days ago \| past \| discuss
		Alibaba open-sources Qwen3.6-35B-A3B, a 35B MoE model with 3B active parameters (huggingface.co)
		10 points by steveharing1 7 days ago \| past \| discuss
		Bonsai 1.7B in the browser: a 290MB 1-bit LLM on WebGPU (huggingface.co)
		8 points by victormustar 8 days ago \| past \| discuss
		HuggingFace launches pre-compiled machine learning kernels repository (huggingface.co)
		3 points by embedding-shape 8 days ago \| past \| 2 comments
		Gemopus: A Gemma fine-tune that prioritizes stability over long chain-of-thought (huggingface.co)
		1 point by steveharing1 9 days ago \| past \| discuss
		Kernels (huggingface.co)
		1 point by tosh 9 days ago \| past \| discuss
		Neo-Unify: An Encoder-Free, Native Multimodal Paradigm (SenseTime) (huggingface.co)
		3 points by WayLonWen 10 days ago \| past \| discuss
		Multimodal Embedding and Reranker Models with Sentence Transformers (huggingface.co)
		1 point by gmays 10 days ago \| past \| discuss
		HuggingFace Papers (huggingface.co)
		2 points by pierre 10 days ago \| past \| discuss
		ALTK‑Evolve: On‑the‑Job Learning for AI Agents (huggingface.co)
		1 point by gmays 10 days ago \| past \| discuss
		Show HN: NRC nuclear licensing RAG pipeline and regulatory embeddings dataset (huggingface.co)
		2 points by davenporten 10 days ago \| past \| discuss
		More

HN For You