Step-3.5-Flash
by StepFun
+
+
+
+
About
Step-3.5-Flash, released by StepFun on February 2, 2026, is a Mixture-of-Experts large language model with 197 billion total parameters and approximately 11 billion active parameters per inference. It features a 256K token context window using a 3:1 sliding-window-to-full-attention ratio, processing 100–350 tokens per second. Step-3.5-Flash targets agentic tasks, coding workflows, and open-source deployments requiring frontier reasoning capabilities with efficient inference, under an Apache 2.0 license.
+
+
+
+
Timeline
ReleasedFeb 2, 2026
+
+
+
+
License & Family
License
Apache 2.0
Performance Overview
Performance metrics and category breakdown
Overall Performance
1 benchmarks
Average Score
74.4%
Best Score
74.4%
High Performers (80%+)
0Top Categories
Coding
74.4%
+
+
+
+
All Benchmark Results for Step-3.5-Flash
Complete list of benchmark scores with detailed information
| SWE Bench Verified | Coding | 74.40 | 74.4% | Unverified |