# MinT: Managed Infrastructure for Training and Serving Millions of LLMs

arXiv: https://arxiv.org/abs/2605.13779 · HTML: https://mlsanity.com/paper/2605.13779/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Mixture of LoRA experts, one of 12 directions in Mixture-of-experts.
- Builds on 29 of its 48 references that are in the map.
- First posted on arXiv 13 May 2026.

## Placed in

- Mixture-of-experts / [Mixture of LoRA experts](https://mlsanity.com/moe/mixture-of-lora-experts/)

## Builds on (most cited first, 29 of 29)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [ZeRO: Memory Optimizations Toward Training Trillion Parameter Models](https://mlsanity.com/paper/1910.02054/) (2019)
- [Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM](https://mlsanity.com/paper/2104.04473/) (2021)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [Kimi K2: Open Agentic Intelligence](https://mlsanity.com/paper/2507.20534/) (2025)
- [Ray: A Distributed Framework for Emerging AI Applications](https://mlsanity.com/paper/1712.05889/) (2017)
- [AdaLoRA: Adaptive Budget Allocation for Parameter-Efficient Fine-Tuning](https://mlsanity.com/paper/2303.10512/) (2023)
- [AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning](https://mlsanity.com/paper/2505.24298/) (2025)
- [S-LoRA: Serving Thousands of Concurrent LoRA Adapters](https://mlsanity.com/paper/2311.03285/) (2023)
- [LawBench: Benchmarking Legal Knowledge of Large Language Models](https://mlsanity.com/paper/2309.16289/) (2023)
- [Punica: Multi-Tenant LoRA Serving](https://mlsanity.com/paper/2310.18547/) (2023)
- [Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers](https://mlsanity.com/paper/2510.11370/) (2025)
- [FinGPT: Democratizing Internet-scale Data for Financial Large Language Models](https://mlsanity.com/paper/2307.10485/) (2023)
- [StreamRL: Scalable, Heterogeneous, and Elastic RL for LLMs with Disaggregated Stream Generation](https://mlsanity.com/paper/2504.15930/) (2025)
- [Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model](https://mlsanity.com/paper/2510.18855/) (2025)
- [FinEval: A Chinese Financial Domain Knowledge Evaluation Benchmark for Large Language Models](https://mlsanity.com/paper/2308.09975/) (2023)
- [NeMo-Aligner: Scalable Toolkit for Efficient Model Alignment](https://mlsanity.com/paper/2405.01481/) (2024)
- [AsyncFlow: An Asynchronous Streaming RL Framework for Efficient LLM Post-Training](https://mlsanity.com/paper/2507.01663/) (2025)
- [Laminar: A Scalable Asynchronous RL Post-Training Framework](https://mlsanity.com/paper/2510.12633/) (2025)
- [Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead](https://mlsanity.com/paper/2407.00066/) (2024)
- [MoE Parallel Folding: Heterogeneous Parallelism Mappings for Efficient Large-Scale MoE Model Training with Megatron Core](https://mlsanity.com/paper/2504.14960/) (2025)
- [mLoRA: Fine-Tuning LoRA Adapters via Highly-Efficient Pipeline Parallelism in Multiple GPUs](https://mlsanity.com/paper/2312.02515/) (2023)
- [Jet-RL: Enabling On-Policy FP8 Reinforcement Learning with Unified Training and Rollout Precision Flow](https://mlsanity.com/paper/2601.14243/) (2026)
- [Improving the Serving Performance of Multi-LoRA Large Language Models via Efficient LoRA and KV Cache Management](https://mlsanity.com/paper/2505.03756/) (2025)
- [Serving Heterogeneous LoRA Adapters in Distributed LLM Inference Systems](https://mlsanity.com/paper/2511.22880/) (2025)

## Built on by (most cited first, 1 of 1)

- [JoyNexus: Service-Oriented Multi-Tenant Post-Training for VLA Models](https://mlsanity.com/paper/2607.16074/) (2026)

## Abstract (arXiv)

We present MindLab Toolkit (MinT), a managed infrastructure system for Low-Rank Adaptation (LoRA) post-training and online serving. MinT targets a setting where many trained policies are produced over a small number of expensive base-model deployments. Instead of materializing each policy as a merged full checkpoint, MinT keeps the base model resident and moves exported LoRA adapter revisions through rollout, update, export, evaluation, serving, and rollback, hiding distributed training, serving, scheduling, and data movement behind a service interface. MinT scales this path along three axes. Scale Up extends LoRA RL to frontier-scale dense and MoE architectures, including MLA and DSA attention paths, with training and serving validated beyond 1T total parameters. Scale Down moves only the exported LoRA adapter, which can be under 1% of base-model size in rank-1 settings; adapter-only handoff reduces the measured step by 18.3x on a 4B dense model and 2.85x on a 30B MoE, while concurrent multi-policy GRPO shortens wall time by 1.77x and 1.45x without raising peak memory. Scale Out separates durable policy addressability from CPU/GPU working sets: a tensor-parallel deployment supports 10^6-scale addressable catalogs (measured single-engine sweeps through 100K) and thousand-adapter active waves at cluster scale, with cold loading treated as scheduled service work and packed MoE LoRA tensors improving live engine loading by 8.5-8.7x. MinT thus manages million-scale LoRA policy catalogs while training and serving selected adapter revisions over shared 1T-class base models.