The AI landscape continues to evolve rapidly. Today’s roundup features new Hugging Face model uploads spanning MoE architectures, TTS, and image generation, a GitHub darling crossing 44,000 stars, and several limited-time free models expiring soon.
Hugging Face Model Updates
Ten new models were uploaded to Hugging Face, spanning MoE architectures, TTS, image generation, reinforcement learning, and more. Highlights include:
- nex-agi/Nex-N2-Pro (π 14) β A Qwen3.5 MoE-based multimodal model supporting both text and image inputs. One of the most anticipated uploads of the day, showing strong community interest.
- turtle170/NetTinyANN (π 8) β Apache 2.0 licensed tiny neural network model, proving that compact architectures still draw attention in the age of billion-parameter giants.
- akrao9/Boomer-T2I (π₯ 433) β A text-to-image diffusion model using linear attention mechanisms for improved efficiency. Already gaining downloads.
- iamseungpil/metacot-h200-e4-gold-decoy (π 2) β Experimental Meta-CoT (Chain-of-Thought) training checkpoint exploring reasoning enhancements.
- Noah733/xyz-5 β A Qwen3-based TTS model with safetensors format, exploring the intersection of LLMs and speech synthesis.
- FirstPotatoCoder/SmolTTS β A compact TTS model, continuing the trend of small specialized models for specific tasks.
- chess-pre-to-post/rl_C6p5e18_680m_alpha1.000_beta0.296 (π 1) β A reinforcement learning checkpoint for chess, exploring pre-to-post training transfer with specific alpha/beta hyperparameters.
- aixk/vlite3.61-100M β Referencing arXiv:1910.09700, this lightweight vision transformer is optimized for endpoints-compatible deployment.
OpenRouter Model Additions
One new model joined the OpenRouter lineup:
- Qwen/Qwen3.7 Plus β Qwen’s latest flagship with a massive 1,000,000-token context window and a prompt price of just $0.40/M tokens. Ideal for document-grounded reasoning, long-context retrieval, and multi-turn agent workflows. With million-token context becoming increasingly affordable, this model is well positioned for production-grade long-context applications.
GitHub Community Highlights
Five repositories gained significant traction:
-
pewdiepie-archdaemon/odysseus (44,687 stars β, JavaScript) β A self-hosted AI workspace that continues its meteoric rise. Gaining thousands more stars since yesterday’s report, it offers a unified environment for running AI agents, managing prompts, and orchestrating model interactions locally. The self-hosted workspace trend shows no signs of slowing.
-
asz798838958/aBaiAutoplus (1,412 stars, Python) β Multi-platform AI account auto-registration and management tool with protocol-based one-click ChatGPT Plus activation. Strong interest from the Chinese developer community.
-
ClaudioDrews/memory-os (769 stars, Python) β A 7-layer memory operating system for Hermes Agent featuring persistent memory with Qdrant and structured facts. Gaining steady traction as agent memory management becomes a critical infrastructure concern.
-
liyue-aigc/female-portrait-director (506 stars) β A modular Codex Skill for directing and expanding detailed AI female portrait prompts, showing the continued appetite for specialized generative AI tooling.
-
DannyMac180/skills (401 stars, Python) β A curated collection of AI agent skills by Dan McAteer, reflecting the growing ecosystem of reusable agent capabilities.
Limited-Time Free Models
Five models are available for free on OpenRouter for a limited time, with three expiring today (June 4):
| Model | Expires | Context | Prompt / Completion Price |
|---|---|---|---|
| Qwen: Qwen3 30B A3B | Jun 5 (π₯ 1 day left) | 131,072 | $0.09/M / $0.45/M |
| Sao10k: Llama 3 Euryale 70B v2.1 | Jun 5 (π₯ 1 day left) | 8,192 | $1.48/M / $1.48/M |
| NousResearch: Hermes 2 Pro - Llama-3 8B | Jun 5 (π₯ 1 day left) | 8,192 | $0.14/M / $0.14/M |
| Nex AGI: DeepSeek V3.1 Nex N1 | Jun 8 (3 days left) | 131,072 | $0.14/M / $0.50/M |
| Anthropic: Claude Opus 4.6 (Fast) | Jun 29 (24 days left) | 1,000,000 | $30/M / $150/M |
Three models expire tomorrow β developers should grab them while they last. Claude Opus 4.6 Fast remains available until the end of the month with its full million-token context. The Nex AGI DeepSeek V3.1 N1 offers an excellent balance of 131K context and low pricing for those needing a bit more time to evaluate.