ModelsCompareBest forBenchmarksStatusPricingAPI

Step 3.7 Flash

StepFun·Pending·Released 2026-05-29textvision
Context256k
Max output-
Input / 1M-
Output / 1M-
Cached in-
Benchmarks19

Benchmarks

Cited public sources · rank is exact-benchmark, apples-to-apples

AA-LCR#3/463.9%
BrowseComp#11/1575.8%
ClawEval v1.167.1%
DeepSearchQA Accuracy81.7%
DeepSearchQA F192.8%
GDPval-Stirrup1415.8
GDPval-Stirrup Intelligence Index45.8
ResearchRubrics#2/371.7%
SimpleVQA (with tools)79.2%
Terminal-Bench 2.1#15/1759.6%
Toolathlon#3/649.5%
V* (with Python)95.3%
WorldVQA (with visual search)58.1%

Access Step 3.7 Flash and every other model through one endpoint with automatic failover: Respan gateway.