Nemotron 3 Ultra 550b A55b vs StepFun: Step 3.5 Flash
Summary
Nemotron 3 Ultra 550b A55b vs Step 3.5 Flash benchmark comparison: Nemotron 3 Ultra 550b A55b leads on average score with 8.1 vs 2.6. Step 3.5 Flash has the lower benchmark cost at $0.020 vs $0.158. Nemotron 3 Ultra 550b A55b is faster at 15.05s vs 39.03s, with pass rates of 69.8% vs 28.6%.
Recommended model: Nemotron 3 Ultra 550b A55b - It has the best score here (8.1), while responding about 2.6x faster than Step 3.5 Flash.
Step 3.5 FlashStep 3.5 FlashnoneArchived model: this model is no longer updated or tested on new tests.Release: 2026-02-01
Score
8.1Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
2.6Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
Rank
#26
#168
Reliability
9.7First-attempt success score: 10.0 means no retryable target API or rate-limit failures before successful calls; tracked failures lower the score.…
10.0First-attempt success score: 10.0 means no retryable target API or rate-limit failures before successful calls; tracked failures lower the score.…
Consistency
8.8Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
5.7Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
Tests Correct
A test is fully passed only if every run passed for that test.Wrong answer: 7API error: 1Response Time (avg)15.05sResponse Time (max)43.93sResponse Time (total)316.09sA test is fully passed only if every run passed for that test.…
A test is fully passed only if every run passed for that test.API error: 4Did not follow instructions: 1Wrong answer: 1Response Time (avg)39.03sResponse Time (max)114.12sResponse Time (total)312.26sA test is fully passed only if every run passed for that test.…
Attempt pass rate
69.8%Attempt pass rate = passed attempts / total attempts across runs.…
28.6%Attempt pass rate = passed attempts / total attempts across runs.…
Flaky tests
3Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
Total Runs
63Total Runs…
36Total Runs…
Cost per result
0.000Shows the average cost per correct benchmark answer in cents (lower is better).…
0.328Shows the average cost per correct benchmark answer in cents (lower is better).…
Total Cost
$0.158Total Cost (Current Price)…
$0.020Total Cost (Current Price)…
Input Price
$0.500 / 1MInput Price…
$0.090 / 1MInput Price…
Output Price
$2.200 / 1MOutput Price…
$0.300 / 1MOutput Price…
Total Input Tokens
46,813Total Input Tokens…
1,971Total Input Tokens…
Output Tokens
18,002Output Tokens…
64,795Output Tokens…
Reasoning Tokens
53,091Reasoning Tokens…
0Reasoning Tokens…
Response Time (avg)
15.05sResponse Time (avg)…
39.03sResponse Time (avg)…
Response Time (max)
43.93sResponse Time (max)…
114.12sResponse Time (max)…
Response Time (total)
316.09sResponse Time (total)…
312.26sResponse Time (total)…
Generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
10.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
100.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)8.62sResponse Time (max)16.86sResponse Time (total)34.49sA test is fully passed only if every run passed for that test.…
8.62sResponse Time (avg)…
780Total Input Tokens…
835Output Tokens…
1,485Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
10.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
100.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)34.99sResponse Time (max)109.60sResponse Time (total)139.95sA test is fully passed only if every run passed for that test.…
8.4Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
7.4Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
88.9%Attempt pass rate = passed attempts / total attempts across runs.…
1Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.Wrong answer: 1Response Time (avg)26.53sResponse Time (max)31.91sResponse Time (total)79.58sA test is fully passed only if every run passed for that test.…
26.53sResponse Time (avg)…
7,686Total Input Tokens…
2,854Output Tokens…
17,725Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
9.8Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
3.3Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.API error: 1Response Time (avg)0msResponse Time (max)0msResponse Time (total)0msA test is fully passed only if every run passed for that test.…
9.8Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
100.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)43.93sResponse Time (max)43.93sResponse Time (total)43.93sA test is fully passed only if every run passed for that test.…
43.93sResponse Time (avg)…
17,574Total Input Tokens…
1,040Output Tokens…
3,590Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
3.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.API error: 1Response Time (avg)0msResponse Time (max)0msResponse Time (total)0msA test is fully passed only if every run passed for that test.…
10.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
100.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)5.68sResponse Time (max)7.94sResponse Time (total)11.36sA test is fully passed only if every run passed for that test.…
5.68sResponse Time (avg)…
7,989Total Input Tokens…
473Output Tokens…
1,285Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
1.5Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
5.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.API error: 1Response Time (avg)0msResponse Time (max)0msResponse Time (total)0msA test is fully passed only if every run passed for that test.…
3.5Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
4.4Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
33.3%Attempt pass rate = passed attempts / total attempts across runs.…
2Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.Wrong answer: 3Response Time (avg)24.90sResponse Time (max)34.96sResponse Time (total)74.71sA test is fully passed only if every run passed for that test.…
24.90sResponse Time (avg)…
858Total Input Tokens…
11,169Output Tokens…
16,249Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
3.3Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
3.3Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
33.3%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)34.54sResponse Time (max)34.54sResponse Time (total)34.54sA test is fully passed only if every run passed for that test.…
3.7Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
9.5Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.API error: 1Response Time (avg)2.52sResponse Time (max)2.52sResponse Time (total)2.52sA test is fully passed only if every run passed for that test.…
2.52sResponse Time (avg)…
360Total Input Tokens…
70Output Tokens…
235Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
4.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.Did not follow instructions: 1Response Time (avg)14.37sResponse Time (max)14.37sResponse Time (total)14.37sA test is fully passed only if every run passed for that test.…
9.8Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
100.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)6.35sResponse Time (max)9.38sResponse Time (total)12.69sA test is fully passed only if every run passed for that test.…
6.35sResponse Time (avg)…
765Total Input Tokens…
182Output Tokens…
1,243Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
5.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
5.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
50.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)9.30sResponse Time (max)9.30sResponse Time (total)9.30sA test is fully passed only if every run passed for that test.…
5.5Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
9.9Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
33.3%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.Wrong answer: 2Response Time (avg)3.54sResponse Time (max)6.03sResponse Time (total)10.62sA test is fully passed only if every run passed for that test.…
3.54sResponse Time (avg)…
792Total Input Tokens…
771Output Tokens…
2,055Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
0.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
0.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)0msResponse Time (max)0msResponse Time (total)0msA test is fully passed only if every run passed for that test.…
10.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
100.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.No failed answers.Response Time (avg)7.72sResponse Time (max)7.72sResponse Time (total)7.72sA test is fully passed only if every run passed for that test.…
7.72sResponse Time (avg)…
9,781Total Input Tokens…
304Output Tokens…
984Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
3.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.API error: 1Response Time (avg)0msResponse Time (max)0msResponse Time (total)0msA test is fully passed only if every run passed for that test.…
3.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.Wrong answer: 1Response Time (avg)38.47sResponse Time (max)38.47sResponse Time (total)38.47sA test is fully passed only if every run passed for that test.…
38.47sResponse Time (avg)…
228Total Input Tokens…
304Output Tokens…
8,240Reasoning Tokens…
Step 3.5 FlashArchived model: this model is no longer updated or tested on new tests.
3.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
0.0%Attempt pass rate = passed attempts / total attempts across runs.…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
A test is fully passed only if every run passed for that test.Wrong answer: 1Response Time (avg)114.12sResponse Time (max)114.12sResponse Time (total)114.12sA test is fully passed only if every run passed for that test.…