MCP servers tagged dpo
10 live MCP servers tagged "dpo", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with dpo.
Safety Training
DPO/SafeDPO/OPAD training + eval for teaching tool-using LLMs to refuse falsely-benign MCP exploits
Dpolens
Give your AI coding assistant access to your organisation's privacy policies and the law, with exact, versioned citations. A self-hosted MCP server.
LLM Dataset Studio
Local-first text and LLM dataset curation studio for JSONL, SFT, DPO, conversations, agent traces, quality review, versioning, exports, MCP, and modular AI plugins.
Evoflow
EvoFlow: 三层进化栈的自进化 LLM Agent 框架 / Self-evolving LLM agent framework (three-layer evolution stack)
Symposium
Multi-agent reasoning MCP server for Claude Code. Spawns parallel research agents to find knowledge LLMs don't have. +23.1% on Terminal Bench 2.0 SWE tasks.
Obsidian Chromadb Self Evolving Kb
面向 Markdown 知识库的自进化 RAG 检索引擎：BGE 向量 + BM25 + RRF 融合检索，两阶段 Rerank，反馈驱动 BGE 微调闭环（离线闸门保障安全上线）。开箱自证检索质量，支持 Obsidian 同步与 FastAPI/MCP 服务。
Dataset Genie
Turn a two-sentence domain brief into an Unsloth-ready fine-tuning dataset. Eight-stage OpenRouter pipeline with LLM judge, human review and SFT / Alpaca / DPO / tool-calling / GR…
Knovaryn
Open, MCP-native training-data foundry: turn permitted documents into traceable, quality-gated SFT and preference datasets.
Persona Forge
Fine-tune persona chatbots that never break character — synthetic roleplay data with judge filtering, DPO against assistant boilerplate, QLoRA recipes, and a local RAG loremaster…
AgentGym
A verifiable Kubernetes incident gym for post-training and serving tool-using agents. Real control plane, deterministic workload simulator, executable rewards.