Submissions from huggingface.co

		Multi-vector Grep for code agents, save 15% tokens, local (huggingface.co)
		3 points by raphaelty 14 hours ago \| past \| 1 comment
		Can LLMs Play the Game of Science? (huggingface.co)
		1 point by amai 1 day ago \| past \| discuss
		Transformers.js v4 Preview: Now Available on NPM (huggingface.co)
		1 point by ibobev 1 day ago \| past \| discuss
		From Golden Gate Bridge to JSON: Why Anthropic's SAE Failed on JSON Output (huggingface.co)
		2 points by maziyar 2 days ago \| past \| 1 comment
		The Future of the Global Open-Source AI Ecosystem: From DeepSeek to AI+ (huggingface.co)
		2 points by mariuz 2 days ago \| past \| discuss
		LingBot – open weights world model (huggingface.co)
		1 point by nikhizzle 2 days ago \| past \| discuss
		MechaEpstein-8000 (huggingface.co)
		1 point by aortega 2 days ago \| past \| 1 comment
		Show HN: Insurance AI Benchmark – 510 scenarios from production (huggingface.co)
		1 point by pavelsukhachev 3 days ago \| past \| discuss
		PaperBanana: Automating Academic Illustration for AI Scientists (huggingface.co)
		2 points by 7777777phil 3 days ago \| past \| discuss
		Sanskrit AI beats CleanRL SOTA by 125% (huggingface.co)
		1 point by prabhatkr 5 days ago \| past \| 1 comment
		The Future of the Global Open-Source AI Ecosystem: From DeepSeek to AI+ (huggingface.co)
		6 points by gmays 5 days ago \| past \| discuss
		Benchmarking how well LLMs can play FizzBuzz (huggingface.co)
		2 points by _venkatasg 6 days ago \| past \| 1 comment
		I trained a 135M TTS model for ~$100, runs 20× real-time on CPU (huggingface.co)
		4 points by sammyyyyyyy 6 days ago \| past \| 2 comments
		Craft – image models can think like LLMs (huggingface.co)
		5 points by denti 7 days ago \| past \| 1 comment
		Sopro v1.5: A 135M TTS model trained for ~$100, runs 20× real-time on CPU (huggingface.co)
		5 points by sammyyyyyyy 7 days ago \| past \| 1 comment
		Show HN: DeepBrainz-R1 – Reasoning-First Small Models for Agentic Systems (huggingface.co)
		3 points by DeepBrainz 7 days ago \| past \| discuss
		The Future of the Global Open-Source AI Ecosystem: From DeepSeek to AI+ (huggingface.co)
		1 point by myk-e 9 days ago \| past \| discuss
		Qwen3-coder-next: SOTA open source coding model (huggingface.co)
		3 points by binsquare 9 days ago \| past \| discuss
		Zai/GLM-OCR (huggingface.co)
		2 points by naze 10 days ago \| past \| 1 comment
		Step 3.5 Flash LLM model, agentic coding ~18x faster than GLM 4.7 / Kimi K2.5 (huggingface.co)
		3 points by skhameneh 10 days ago \| past \| 1 comment
		Post anything on Moltbook as a human, without AI (huggingface.co)
		2 points by shash42 10 days ago \| past \| 1 comment
		One Year Since the "DeepSeek Moment" (huggingface.co)
		3 points by myk-e 11 days ago \| past \| discuss
		VibeVoice-ASR: speech-to-text model designed to handle 60-minute long-form audio (huggingface.co)
		11 points by maxloh 12 days ago \| past \| discuss
		Nvidia Nemotron 3-Nano 30B-A3B-BF16 (huggingface.co)
		10 points by tosh 12 days ago \| past \| 1 comment
		From HashHop to Memory-Augmented Language Models (huggingface.co)
		2 points by codelion 13 days ago \| past \| discuss
		Show HN: SOTA NLP Models (huggingface.co)
		1 point by ChadNauseam 13 days ago \| past \| discuss
		Daggr: Chain apps programmatically, inspect visually (huggingface.co)
		2 points by ibobev 14 days ago \| past
		Nvidia Cosmos Policy for Advanced Robot Control (huggingface.co)
		1 point by ibobev 14 days ago \| past
		LongCat-Flash-Lite 100B A3B Technical Report [pdf] (huggingface.co)
		2 points by limoce 14 days ago \| past
		We got Claude to teach open models how to write CUDA kernels (huggingface.co)
		3 points by mooreds 14 days ago \| past
		More