# Beyond the Trial-and-Error Loop: Hybrid Projection and Automated Tuning for Distributed Training

arXiv: https://arxiv.org/abs/2610.19423 · HTML: https://mlsanity.com/paper/2610.19423/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 33 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Builds on (most cited first, 17 of 17)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM](https://mlsanity.com/paper/2104.04473/) (2021)
- [FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision](https://mlsanity.com/paper/2407.08608/) (2024)
- [MegaBlocks: Efficient Sparse Training with Mixture-of-Experts](https://mlsanity.com/paper/2211.15841/) (2022)
- [Beyond Data and Model Parallelism for Deep Neural Networks](https://mlsanity.com/paper/1807.05358/) (2018)
- [Large Language Models to Enhance Bayesian Optimization](https://mlsanity.com/paper/2402.03921/) (2024)
- [Large Language Model Agent for Hyper-Parameter Optimization](https://mlsanity.com/paper/2402.01881/) (2024)
- [Galvatron: Efficient Transformer Training over Multiple GPUs Using Automatic Parallelism](https://mlsanity.com/paper/2211.13878/) (2022)
- [Improving Automatic Parallel Training via Balanced Memory Workload Optimization](https://mlsanity.com/paper/2307.02031/) (2023)
- [Proteus: Simulating the Performance of Distributed DNN Training](https://mlsanity.com/paper/2306.02267/) (2023)
- [Lumos: Efficient Performance Modeling and Estimation for Large-scale LLM Training](https://mlsanity.com/paper/2504.09307/) (2025)
- [tritonBLAS: Triton-based Analytical Approach for GEMM Kernel Parameter Selection](https://mlsanity.com/paper/2512.04226/) (2025)

## Abstract (arXiv)

Configuring distributed LLM training across hundreds of GPUs requires jointly optimizing tensor, pipeline, expert, context, and data parallelism alongside micro-batch size, recomputation, sharding, and overlap, a combinatorial space where memory legality can hinge on a single byte of activation accounting. Today, teams navigate it through expensive multi-node trial and error, burning GPU-hours on crashes and suboptimal recipes. We present Primus Projection, which closes this loop with two coupled components. First, a hybrid projection engine that measures what it can and simulates what it cannot: GPU microbenchmarks on as few as a single device capture compute kernels and any in-scope communication; a sub-node downscale-measure-upscale workflow analytically restores out-of-scope collectives, FSDP overlap, and the full pipeline schedule at the multi-node target. A pure simulation mode runs entirely on CPU, enabling pre-silicon planning without any GPU. Second, the Primus Tuning Agent treats the projection engine as a fast scoring oracle (seconds per candidate instead of tens of minutes) and uses it to drive an intelligent search over the configuration space. A deterministic seed planner sweeps single axes; an LLM investigation loop proposes cross-axis combinations the planner cannot reach, with an analytic memory pre-filter that rejects infeasible candidates before any GPU time is spent. Validated on Llama 3.1 (dense) and Mixtral 8x22B (MoE) across two GPU generations, all projections stay within ~10% of measured multi-node throughput, and fidelity holds at production scale: 4.7% error on DeepSeek V3 671B at 1,024 GPUs and 1.9% on Qwen3-235B-A22B at 256 GPUs. In a Mixtral 8x22B case study, the Primus Tuning Agent discovers a configuration delivering +27% throughput over the published 4-node BF16 reference at the same scale, in under 30 minutes of single-node exploration.