Recent work has begun to equip vision-language-action (VLA) policies with explicit intermediate reasoning. In embodied control, however, textual chain-of-thought is a poor fit: irrelevant or weakly textual information can interfere with action prediction, while autoregressive text decoding adds too much latency for real-time closed-loop execution. We present VISUALTHINK-VLA, a visual intermediate-reasoning framework for accurate, low-latency VLA policies. We define visual intermediate reasoning operationally as constructing typed visual evidence, selecting a task-dependent budgeted route, and composing the selected channels into compact states before action decoding. Our bootstrapping philosophy is to guide action with effective visual thinking: VISUALTHINK-VLA bootstraps action prediction through a compact visual-evidence interface that preserves spatial precision and supports low-overhead decoding. Besides, to further improve performance and efficiency, VISUALTHINK-VLA adopts a tailored selective routing mechanism to learn the visual evidence tokens, enabling low-latency inference while preserving high-capacity specialization. We also introduce VisualEvidence-Kit, a supervision-and-audit resource centered on a VisualEvidence-Agent that constructs 754.7k VLA instructions in the VisualEvidence-Set for route supervision and counterfactual faithfulness tests. Across multiple benchmarks and real-robot evaluation, VISUALTHINK-VLA achieves the highest success rate on most benchmarks while reducing the multi-second latency of reasoning-augmented baselines to the sub-second regime. On BridgeData V2, policy-side latency falls from 8.377,s with ECoT to 0.367,s, achieving a 22.8× speedup.

While Parameter-Efficient Fine-Tuning (PEFT) methods like Low-Rank Adaptation (LoRA) effectively address resource constraints during fine-tuning, their performance often falls short, especially in multidimensional task scenarios. To address this issue, one straightforward solution is to introduce task-specific LoRA as domain experts, leveraging the modeling of multiple capabilities of experts and thus enhancing the general capability of multi-task learning.Although promising, these additional components often add complexity to the training and inference process, contravening the efficiency that PEFT is designed to deliver. Considering this, we introduce an innovative PEFT method, TeamLoRA, consisting of a collaboration and competition module for LoRA experts, thus achieving the right balance of effectiveness and efficiency:(i) For collaboration, we introduce a novel knowledge sharing and organization mechanism designed to optimize hierarchical learning while enhancing the efficiency of model training and inference.(ii) For competition, we propose leveraging a game-theoretic interaction mechanism for experts, encouraging experts to transfer their domain-specific knowledge while facing diverse downstream tasks, thus enhancing the performance.By doing so, TeamLoRA elegantly connects the experts as a “Team” with internal collaboration and competition, enabling a faster and more accurate PEFT paradigm. Meanwhile, we curate a Comprehensive Multi-Task Evaluation (CME) benchmark to thoroughly assess the capability of multi-task learning. Experiments conducted on our CME and other benchmarks indicate the effectiveness and efficiency of TeamLoRA. Our project is available at https://github.com/DCDmllm/TeamLoRA.