StepFun: Step 3.5 Flash

Step 3.5 Flash is a text-in, text-out model from StepFun with a 262,144-token context window and a 16,384-token output limit. It supports tool use and reasoning, which makes it usable for multi-step tasks and agentic workflows. Structured output support is unconfirmed. It is a paid model with no indication of open weights. At $0.09 per million input tokens and $0.30 per million output tokens, it sits in the budget tier, making cost the clearest argument for choosing it. Its blended benchmark score of 42.4 comes from a single benchmark, so quality comparisons are thin and should be treated cautiously. Buyers who prioritize low inference cost for high-volume, reasoning-capable workloads may find it worth testing, but those who need confidence from broad third-party evaluation should note that the benchmark coverage here is minimal.