TranslatedBullFlow — Open-source enterprise RAG solution: knowledge base → automated marketing content → omnichannel publishing. The entire knowledge base pipeline runs on local open-source models, with no paid models needed; AI Agent creates image-text posts / images / videos, with one-click publishing to 17 platforms including Douyin, Xiaohongshu, WeChat Official Accounts, LinkedIn, and YouTube. Commercial-grade, multi-tenant, billing & admin built in. MIT
Xolotl: run DeepSeek-V4.1-Flash, all 552B, as a local coding agent on one gaming PC (RTX 4090 + 110 GB RAM): ~33 tok/s on agent tasks, ~41 in fast mode, 512K context. Plus GLM-5.3-Flash, Qwen3.8 and LFM3-Omni. MoE experts tiered across VRAM, pinned RAM and NVMe, with Xolotl Direct NVMe->VRAM. OpenAI + Anthropic API.
WeVisDoc is an end-to-end document parser for page images. Fine-tuned from Qwen3-VL-2B-Instruct and Qwen3-VL-4B-Instruct, it turns a page into structured Markdown, with LaTeX formulas and HTML tables.
DeepSeek-V4.1-Flash-1017 EXL3 on 2x DGX Sparks with TensorFold
Fast fork of llama.cpp designed for Qwen3.8-Flash-Next
A Unified Framework for Efficient RSI Research
- agent
- recursive-self-improvement
Calibrain Harness: a forward-testing harness for trading crypto perpetuals with fast classification models (Kairos today) on Bybit testnet. Under development.
C++/CUDA single-GPU inference engine for Qwen3.8 (incl. Qwen3.8-Flash-Next with offloaded experts) on the RTX 5090; grew from NInfer
Reproducible Docker/vLLM deployment, tuning, and benchmarks for DeepSeek V4.1 Flash on a switchless three-node or four-node DGX Spark fabric.
PROMETHEVS: run the full DeepSeek-V4.1-Flash-1017 (510 GB) on a gaming PC with 64 GB RAM and a 12 GB GPU. PROMETHEVS engine (compress-then-correct), OpenAI-compatible local server, report and logs.
Investor’s Last Exam: Can AI Agents Do Equity Investment Without Seeing the Future?
GLM 5.3 Abliterated or Uncensored AI model — 35 jailbreak prompts, model guide and API. One 7API key also covers Claude, GPT, Gemini, Grok, DeepSeek and Kimi.
- abliterated
- abliterated-models
- anthropic-compatible
- claude-code
JevStyle: local decision models for AI agents. Classify, route and score with a probability per option. systemone-compatible local server, MCP tools, Claude Code guard. Open weights, Apache-2.0.
BLA blinds your agent. LASIK gives it its sight back. Static de-blinding CLI + MCP server that undoes BLA (Blind LLM Agents) hardening so coding agents and REA can read the real code again. Never executes the target.
Curated Gemini Omni prompts, API guide, and video examples — video generation, editing, camera control, style transfer, text rendering, and multi-modal workflows.
- ai-prompts
- deepmind
- gemini
- gemini-omni
ReAlign models instantly, on demand. Per-request alignment for verified users and channels, or Always On. Use directly or with Hermes, OpenClaw and custom harnesses.