Category · updated 30 October
New AI tools launched today
24 of the top 100 launches on 30 October 2026 are AI tools, from Hugging Face, App Store, Show HN, GitHub and Launch HN. 89 have ranked since 14 October 2026, newest first. The ranking is rebuilt every morning by 09:00 UTC.
- 3
Nanbeige4.3 33BOpen source
Text generation model by Nanbeige
Trending on Hugging Face · trending score 768, 783 likes - 5
Muse from MetaFree
Muse is your personal AI agent that gets things done, always working on what matters to you.
New app, #2 in Top Free - 6
DeepSeek V4.1 Flash 1017Open source
DeepSeek-V4.1-Flash-1017 is a minor update to DeepSeek-V4.1-Flash, delivering substantial improvements…
Trending on Hugging Face · trending score 864, 3.9k likes - 7
LFM3 1.7BOpen source
LFM3-1.7B is a text model for local agents and specialized applications, based on our new tri-hybrid…
Trending on Hugging Face · trending score 492, 502 likes - 9
FlintOpen source
Flint is Recursive’s mixture-of-experts language model for coding agents, with 39.8 billion total…
Trending on Hugging Face · trending score 455, 1.2k likes - 12
DeepSeek V4.1 Flash 1017 GSQ RCO GGUFOpen source
DeepSeek-V4.1-Flash-1017 is a sparse mixture-of-experts model with 763.2B parameters: 384 routed experts…
Trending on Hugging Face · trending score 427, 481 likes - 15
NVIDIA Nemotron 3.5 Transcribe 30B A4BOpen source
NVIDIA Nemotron-3.5-Transcribe is an offline speech and audio model that processes recordings, together…
Trending on Hugging Face · trending score 408, 429 likes - 16
Doomformer compiles integer neural networks into ZDoom maps and runs them in ViZDoom.
#2 on Show HN yesterday · 72 points - 19
Unlimited OCR 2Open source
Image text to text model by baidu
Trending on Hugging Face · trending score 386, 425 likes - 20
Unlock FP/BF16 instruction issue rate on Nvidia 3000's cards to their professional equivalents
632 stars on GitHub today - 23
Spark X2.6 4BOpen source
Text generation model by XHToken
Trending on Hugging Face · trending score 359, 912 likes - 27
Mistral Large 4 1T A52BOpen source
Mistral Large 4 is our largest and most capable model to date: a 1 trillion-parameter natively multimodal…
Trending on Hugging Face · trending score 321, 1.2k likes - 31
Flux 3 Image Klein ComfyOpen source
Image to image model by black-forest-labs
Trending on Hugging Face · trending score 350, 2.1k likes - 35
learnharnessOpen source
LearnHarness uses AI agents to plan and teach lessons from your current understanding to your goal.
391 stars on GitHub today - 36
Run, inspect, and test real firmware on simulated hardware, from a single MCU to connected devices. Learn…
Launch HN from YC F26 · 64 points - 45
MiniMax M3.1 FlashOpen source
MiniMax-M3.1-Flash is a native multimodal mixture-of-experts model for coding, agentic workflows, computer…
Trending on Hugging Face · trending score 309, 492 likes - 53
OpenDuetOpen source
An Open Source template for a proactive AI agent that gets your work done and keeps you in the loop.
167 stars on GitHub today - 56
Qwen3.5 4B 3D CoderOpen source
An experiment exploring the feasibility of small language models for general-purpose coding and…
Trending on Hugging Face · trending score 274, 290 likes - 66
Simready CentralOpen source
Semantic search for 3D assets: USD + RTX simulation previews
Trending on Hugging Face · trending score 217, 283 likes - 67
DS-BotOpen source
A team of long-lived bots for DeepSeek Harness, with main bots, bot-to-bot messages and group chats.
117 stars on GitHub today - 76
NVIDIA Nemotron 3.5 Bolt E4B 4GBOpen source
NVIDIA-Nemotron-3.5-Bolt-E4B-4GB is a small language model (SLM) further post-trained from Gemma-4-E4B-it…
Trending on Hugging Face · trending score 244, 261 likes - 79
LFM3 Omni 2BOpen source
LFM3-Omni-2B is Liquid AI's first omni-modal model in the LFM3 family: a single end-to-end model that…
Trending on Hugging Face · trending score 209, 214 likes - 95
MOSS-ImageOpen source
MOSS-Image by OpenMOSS: text-to-image generation with native text-driven layout control. Model weights,…
76 stars on GitHub today - 99
AudayPaid
Turn your Apple Watch into your all-day AI recorder. Capture the words and the life around them.
New app, #76 in Top Paid - 106
- 122
RSI-AnythingOpen source
RSI-Anything is a domain-general framework that turns world-model predictions and interaction feedback into recursive self-improvement.
51 stars on GitHub today - 133
GEOInflowOpen source
Self-hosted web and AI crawler analytics for Generative Engine Optimization (GEO). Track visitors, AI…
43 stars on GitHub today - 153
Drop-in FSR 3.1 upscaler for Direct3D 12 that runs DLSS's network (Model E) on AMD GPUs
33 stars on GitHub today - 13
Hy SimulMT 30B A3B FP8Open source
Hy-SimulMT is a real-time simultaneous translation model developed by Tencent Hunyuan, supporting…
Trending on Hugging Face · trending score 422, 429 likes - 63
Carbon XOpen source
An open 14B language model for DNA understanding and generation, from the Carbon family.
Trending on Hugging Face · trending score 249, 261 likes - 90
MOSS Image SFTOpen source
MOSS-Image-SFT is a supervised fine-tuned text-to-image model in the MOSS-Image family from OpenMOSS,…
Trending on Hugging Face · trending score 225, 819 likes - 28
Sage A1.5Open source
Sage‑A1.5 is a 35B Mixture‑of‑Experts agentic model based on Qwen3.6 and Agents-A1 from Shanghai AI…
Trending on Hugging Face · trending score 328, 347 likes - 40
Flint GGUFOpen source
Flint is Recursive’s mixture-of-experts language model for coding agents, with 39.8 billion total…
Trending on Hugging Face · trending score 283, 300 likes - 63
Drama 3Open source
Fish Audio Drama 3 is Fish Audio's new-generation text-to-speech model, built for multi-character scenes…
Trending on Hugging Face · trending score 238, 1k likes - 74
- 75
- 83
Raon 1 Mini PreviewOpen source
Raon-1 mini is a 30B-A4B Mixture-of-Experts (MoE) language model, designed and trained from scratch by…
Trending on Hugging Face · trending score 187, 202 likes - 60
textfoldOpen source
Fold papers, web pages, files, and AI replies into five levels of detail. Zoom from an overview to the…
98 stars on GitHub today - 69
KromossOpen source
Kromoss is a full fine-tune of MOSS-Image-Base at 1024 px.
Trending on Hugging Face · trending score 238, 263 likes - 84
Timestral 209MOpen source
Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M,…
Trending on Hugging Face · trending score 227, 376 likes - 99
MOSS Image BaseOpen source
MOSS-Image-Base is the base text-to-image model in the MOSS-Image family from OpenMOSS.
Trending on Hugging Face · trending score 221, 556 likes - 46
DeepSeek-BotOpen source
A team of long-lived bots for DeepSeek Harness, with main bots, bot-to-bot messages and group chats.
181 stars on GitHub today - 72
Embeddinggemma 2Open source
EmbeddingGemma 2 is an open multimodal embedding model built by Google DeepMind which maps text (incl.…
Trending on Hugging Face · trending score 227, 2.9k likes - 91
StartLux-DecisionOpen source
Open decision models from 0.8B to 27B: a calibrated probability for every option in one forward pass, and…
61 stars on GitHub today - 56
HumanizerOpen source
A 12B model that rewrites AI-written drafts (emails, essays, reports, forum posts; English and Chinese) so…
Trending on Hugging Face · trending score 241, 2.1k likes - 71
Indra OcrOpen source
Indra OCR is a 1M-parameter OCR model that runs on CPU.
Trending on Hugging Face · trending score 236, 394 likes - 99
timestralOpen source
Repository for Timestral, the time-series foundation model by Mistral AI
36 stars on GitHub today - 23
- 71
Muse Spark Open 816B A42BOpen source
Image text to text model by meta-models
Trending on Hugging Face · trending score 255, 669 likes - 84
Numbers are each launch's rank on its most recent day. These are the newest 50 of 89; older ones are in every daily ranking. See the full ranking