Back

StepFun

Step 3.5 Flash

ReasoningOpen weights
Intelligence
16.6
Rank 292 of 747 measured models
Output speed
127 t/s
Rank 127 of 405 measured models
Price/1M
$0.15
Rank 129 of 664 measured models
Context window
256K
Rank 521 of 1207 measured models

Step 3.5 Flash scores 16.6 on the Artificial Analysis Intelligence Index, ranking 292nd of 747 measured models. The closest models on this index are Gemma 4 26B A4B (16.7), Claude 4 Opus (16.6), and Ring-2.6-1T (16.6). At $0.15 per million tokens (3:1 input/output blended price), it is cheaper than Gemma 4 26B A4B ($0.175), Claude 4 Opus ($30), and Ring-2.6-1T ($0.85). With an output speed of 127 t/s, it is faster than Gemma 4 26B A4B (62.7 t/s) and Ring-2.6-1T (117 t/s).

Artificial Analysis Indices
Intelligence16.6
Rank 292 of 747 measured models
Standard Benchmarks
GPQA Diamond83.1%
Rank 171 of 620 measured models
HLE21.1%
Rank 216 of 667 measured models
IFBench64.6%
Rank 108 of 450 measured models
AA-LCR50.0%
Rank 329 of 578 measured models
TerminalBench Hard27.3%
Rank 132 of 432 measured models
τ²-Bench94.4%
Rank 33 of 440 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy