Gemini 3.5 Flash vs Grok 4.20 Beta benchmark comparison: Gemini 3.5 Flash leads on average score with 9.6 vs 8.5. Grok 4.20 Beta has the lower benchmark cost at $0.750 vs $1.115. Gemini 3.5 Flash is faster at 8.84s vs 9.75s, with pass rates of 96.8% vs 81.5%.
Recommended model: Gemini 3.5 Flash - It has the strongest score in this comparison (9.6) and the best overall balance of cost and response time across all 2 models.
Grok 4.20 BetaGrok 4.20 BetamediumArchived model: this model is no longer updated or tested on new tests.Release: 2026-03-12
Score
9.6Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
8.5Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
Rank
#2
#14
Reliability
10.0First-attempt success score: 10.0 means no retryable target API or rate-limit failures before successful calls; tracked failures lower the score.…
N/AFirst-attempt success score: 10.0 means no retryable target API or rate-limit failures before successful calls; tracked failures lower the score.…
Consistency
9.6Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
9.5Consistency 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: 1Response Time (avg)8.84sResponse Time (max)34.82sResponse Time (total)185.57sA 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.Wrong answer: 3Did not follow instructions: 1Response Time (avg)9.75sResponse Time (max)31.36sResponse Time (total)175.48sA test is fully passed only if every run passed for that test.…
Attempt pass rate
96.8%Attempt pass rate = passed attempts / total attempts across runs.…
81.5%Attempt pass rate = passed attempts / total attempts across runs.…
Flaky tests
1Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
1Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
Total Runs
63Total Runs…
52Total Runs…
Cost per result
5.575Shows the average cost per correct benchmark answer in cents (lower is better).…
4.505Shows the average cost per correct benchmark answer in cents (lower is better).…
Total Cost
$1.115Total Cost (Current Price)…
$0.750Total Cost (Current Price)…
Input Price
$1.500 / 1MInput Price…
$5.805 / 1MInput Price…
Output Price
$9.000 / 1MOutput Price…
$5.805 / 1MOutput Price…
Total Input Tokens
37,594Total Input Tokens…
35,955Total Input Tokens…
Output Tokens
1,975Output Tokens…
1,647Output Tokens…
Reasoning Tokens
115,638Reasoning Tokens…
91,565Reasoning Tokens…
Response Time (avg)
8.84sResponse Time (avg)…
9.75sResponse Time (avg)…
Response Time (max)
34.82sResponse Time (max)…
31.36sResponse Time (max)…
Response Time (total)
185.57sResponse Time (total)…
175.48sResponse 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)2.57sResponse Time (max)3.60sResponse Time (total)10.27sA test is fully passed only if every run passed for that test.…
2.57sResponse Time (avg)…
492Total Input Tokens…
174Output Tokens…
4,997Reasoning Tokens…
Grok 4.20 BetaArchived model: this model is no longer updated or tested on new tests.
8.7Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
7.9Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
91.7%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)3.16sResponse Time (max)3.44sResponse Time (total)12.65sA 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)22.96sResponse Time (max)34.82sResponse Time (total)68.88sA test is fully passed only if every run passed for that test.…
22.96sResponse Time (avg)…
8,118Total Input Tokens…
456Output Tokens…
47,129Reasoning Tokens…
Grok 4.20 BetaArchived 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)31.36sResponse Time (max)31.36sResponse Time (total)31.36sA 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)22.37sResponse Time (max)22.37sResponse Time (total)22.37sA test is fully passed only if every run passed for that test.…
22.37sResponse Time (avg)…
12,873Total Input Tokens…
351Output Tokens…
16,323Reasoning Tokens…
Grok 4.20 BetaArchived 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)20.93sResponse Time (max)20.93sResponse Time (total)20.93sA 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)6.43sResponse Time (max)8.51sResponse Time (total)12.87sA test is fully passed only if every run passed for that test.…
6.43sResponse Time (avg)…
7,548Total Input Tokens…
279Output Tokens…
8,466Reasoning Tokens…
Grok 4.20 BetaArchived 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)4.01sResponse Time (max)4.27sResponse Time (total)8.02sA test is fully passed only if every run passed for that test.…
7.6Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
7.2Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
77.8%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)14.09sResponse Time (max)22.00sResponse Time (total)42.27sA test is fully passed only if every run passed for that test.…
14.09sResponse Time (avg)…
633Total Input Tokens…
12Output Tokens…
24,721Reasoning Tokens…
Grok 4.20 BetaArchived model: this model is no longer updated or tested on new tests.
5.3Summarizes 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).…
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)21.33sResponse Time (max)24.21sResponse Time (total)64.00sA 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)3.63sResponse Time (max)3.63sResponse Time (total)3.63sA test is fully passed only if every run passed for that test.…
3.63sResponse Time (avg)…
486Total Input Tokens…
115Output Tokens…
1,650Reasoning Tokens…
Grok 4.20 BetaArchived 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)5.78sResponse Time (max)5.78sResponse Time (total)5.78sA 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)3.35sResponse Time (max)3.42sResponse Time (total)6.69sA test is fully passed only if every run passed for that test.…
3.35sResponse Time (avg)…
615Total Input Tokens…
70Output Tokens…
3,799Reasoning Tokens…
Grok 4.20 BetaArchived 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.…
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)4.89sResponse Time (max)5.89sResponse Time (total)9.78sA 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)3.23sResponse Time (max)3.68sResponse Time (total)9.69sA test is fully passed only if every run passed for that test.…
3.23sResponse Time (avg)…
558Total Input Tokens…
241Output Tokens…
4,940Reasoning Tokens…
Grok 4.20 BetaArchived 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)3.52sResponse Time (max)4.53sResponse Time (total)10.57sA 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)4.96sResponse Time (max)4.96sResponse Time (total)4.96sA test is fully passed only if every run passed for that test.…
4.96sResponse Time (avg)…
6,115Total Input Tokens…
265Output Tokens…
1,608Reasoning Tokens…
Grok 4.20 BetaArchived 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.Did not follow instructions: 1Response Time (avg)12.39sResponse Time (max)12.39sResponse Time (total)12.39sA 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)3.94sResponse Time (max)3.94sResponse Time (total)3.94sA test is fully passed only if every run passed for that test.…
3.94sResponse Time (avg)…
156Total Input Tokens…
12Output Tokens…
2,005Reasoning Tokens…
Grok 4.20 BetaArchived model: this model is no longer updated or tested on new tests.