AirLLM 70B inference with single 4GB GPU

BELLE: Be Everyone's Large Language model Engine（开源中文对话大模型）

An open-source educational chat model from ICALK, East China Normal University. 开源中英教育对话大模型。(通用基座模型，GPU部署，数据清理) 致敬: LLaMA, MOSS, BELLE, Ziya, vLLM

⚡ Zero-Stall MoE Inference via Lookahead Prediction & Async DMA Prefetching. Optimized for SSD I/O with Hybrid MLA+Sliding Window Attention.

Same model, different wrapper: a from-scratch benchmark comparing coding-agent harnesses (codex, pi, opencode, cursor, devin) and open models on correctness, speed, and token cost

brick is a smart AI Models router, based on complexity & capabilities extraction from the query to the models via proprietary spatial embedding algorythm

Formal Psychological Models of Categorization and Learning

A simple prompt-based approach to detecting prompt injection and jailbreaking attempts using small, self-hosted language models.

Brick of Knowledge on Open Models : Open Source, Open Science, Open Education, Open Collaboration, Open Hardware...

Evidence-backed structural validation of Kimi K3 UD-IQ1_M and UD-Q4_K_XL split GGUF releases using OMIV.

🇫🇷 parler: Multilingual voice intelligence built on Mistral Voxtral model — decision logs from French/English meetings

A framework that enables consistent assessment across environmental claims.

Desktop AI coding-agent platform (Tauri + Rust) for racing, evaluating & orchestrating LLMs across Anthropic and open models (Kimi, MiniMax, DeepSeek, GLM, Qwen) — with a deterministic Axolotl Civilization arena the models compete in.

An independent, evidence-graded read on how much you own the open models and inference providers you rely on.

Systematic scoping review paper for healthcare DES model sharing.

AI-assisted setup kit for testing a low-cost open-model orchestrator/executor stack.

A self-improving agent harness for open models — brain, hands, and super legs for any LLM, frontier or local.

Space and tools for the digital world.

Open-source, locally runnable text-to-video generation — inspired by Sora, built for everyone. Run cinematic video synthesis offline, with full control and transparency.

400+ free LLM models — Llama, Qwen, DeepSeek, Gemma, Mistral, GLM, Kimi, Nemotron, Command A+, MiniMax. Free API access, daily updated. Compare context windows, rate limits, modalities (text, vision, audio, code).