# Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

arXiv: https://arxiv.org/abs/2602.10604 · HTML: https://mlsanity.com/paper/2602.10604/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- First posted on arXiv 11 Feb 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)
- Mixture-of-experts / [MoE in reinforcement learning and robotics](https://mlsanity.com/moe/moe-in-reinforcement-learning-and-robotics/)

## Built on by (most cited first, 30 of 36)

- [Kimi K3: Open Frontier Intelligence](https://mlsanity.com/paper/2607.24653/) (2026)
- [JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting](https://mlsanity.com/paper/2606.18394/) (2026)
- [Prefill-as-a-Service: KVCache of Next-Generation Models Could Go Cross-Datacenter](https://mlsanity.com/paper/2604.15039/) (2026)
- [Symmetry-Compatible Principle for Optimizer Design: Embeddings, LM Heads, SwiGLU MLPs, and MoE Routers](https://mlsanity.com/paper/2605.18106/) (2026)
- [Training with Harnesses: On-Policy Harness Self-Distillation for Complex Reasoning](https://mlsanity.com/paper/2605.08741/) (2026)
- [Seeing with You: Perception-Reasoning Coevolution for Multimodal Reasoning](https://mlsanity.com/paper/2603.28618/) (2026)
- [KAT-Coder-V2 Technical Report](https://mlsanity.com/paper/2603.27703/) (2026)
- [OneReason Technical Report](https://mlsanity.com/paper/2606.06260/) (2026)
- [Loop the Loopies!](https://mlsanity.com/paper/2607.16051/) (2026)
- [FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents](https://mlsanity.com/paper/2606.12087/) (2026)
- [From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI](https://mlsanity.com/paper/2606.14502/) (2026)
- [InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning](https://mlsanity.com/paper/2606.12195/) (2026)
- [Rethinking the Role of Efficient Attention in Hybrid Architectures](https://mlsanity.com/paper/2606.15378/) (2026)
- [Mach-Mind-4-Flash Technical Report](https://mlsanity.com/paper/2607.09375/) (2026)
- [Harness-IF: Evaluating Instruction Following Across Instruction Surfaces in Coding Agents](https://mlsanity.com/paper/2608.11727/) (2026)
- [Consolidating RLVR Capabilities Across Domains: A Deep Dive into Fusion Paradigms](https://mlsanity.com/paper/2608.27409/) (2026)
- [CODA: Difficulty-Aware Compute Allocation for Adaptive Reasoning](https://mlsanity.com/paper/2603.08659/) (2026)
- [QED-Nano: Teaching a Tiny Model to Prove Hard Theorems](https://mlsanity.com/paper/2604.04898/) (2026)
- [The Cancellation Hypothesis in Critic-Free RL: From Outcome Rewards to Token Credits](https://mlsanity.com/paper/2605.08666/) (2026)
- [Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness](https://mlsanity.com/paper/2605.10379/) (2026)
- [Data-Constrained Language Model Pretraining: Improved Regularization and Scaling Laws](https://mlsanity.com/paper/2606.06888/) (2026)
- [SearchSwarm: Towards Delegation Intelligence in Agentic LLMs for Long-Horizon Deep Research](https://mlsanity.com/paper/2606.09730/) (2026)
- [Attention Editing: A Versatile Framework for Cross-Architecture Attention Conversion](https://mlsanity.com/paper/2604.05688/) (2026)
- [CONSCIENTIA: Can LLM Agents Learn to Strategize? Emergent Deception and Trust in a Multi-Agent NYC Simulation](https://mlsanity.com/paper/2604.09746/) (2026)
- [MathDuels: A Self-Play Benchmark That Grows](https://mlsanity.com/paper/2604.21916/) (2026)
- [Beyond Trajectory Imitation: Strategy-Guided Policy Optimization for LLM Reasoning](https://mlsanity.com/paper/2606.24064/) (2026)
- [Enhancing Mixture-of-Experts Specialization via Cluster-Aware Upcycling](https://mlsanity.com/paper/2604.13508/) (2026)
- [Knowledge Index of Noah's Ark](https://mlsanity.com/paper/2606.05104/) (2026)
- [ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping](https://mlsanity.com/paper/2606.31693/) (2026)
- [MOSAIC: Adaptive Inter-layer Composition for Efficient Heterogeneous Vision-Language Models](https://mlsanity.com/paper/2607.09029/) (2026)

## Abstract (arXiv)

We introduce Step 3.5 Flash, a sparse Mixture-of-Experts (MoE) model that bridges frontier-level agentic intelligence and computational efficiency. We focus on what matters most when building agents: sharp reasoning and fast, reliable execution. Step 3.5 Flash pairs a 196B-parameter foundation with 11B active parameters for efficient inference. It is optimized with interleaved 3:1 sliding-window/full attention and Multi-Token Prediction (MTP-3) to reduce the latency and cost of multi-round agentic interactions. To reach frontier-level intelligence, we design a scalable reinforcement learning framework that combines verifiable signals with preference feedback, while remaining stable under large-scale off-policy training, enabling consistent self-improvement across mathematics, code, and tool use. Step 3.5 Flash demonstrates strong performance across agent, coding, and math tasks, achieving 85.4% on IMO-AnswerBench, 86.4% on LiveCodeBench-v6 (2024.08-2025.05), 88.2% on tau2-Bench, 69.0% on BrowseComp (with context management), and 51.0% on Terminal-Bench 2.0, comparable to frontier models such as GPT-5.2 xHigh and Gemini 3.0 Pro. By redefining the efficiency frontier, Step 3.5 Flash provides a high-density foundation for deploying sophisticated agents in real-world industrial environments.