Claude Opus 4.6 (medium) vs Claude Sonnet 4.6 (medium) vs GPT-5.3-Codex (medium) vs Gemini 3.1 Pro Preview (medium) benchmark comparison:GPT-5.3-Codex (medium) leads on Score with 9.1. Claude Opus 4.6 (medium) leads on Reliability with 10.0. GPT-5.3-Codex (medium) has the lowest Total Cost at $1.318. GPT-5.3-Codex (medium) is fastest at 18.87s.
Last updated at:
2026-10-06
Compared models
Rank
#228
Total Output Tokens
96,722
Response Time (avg)
32.23s
Total Cost
$2.961
Rank
#212
Total Output Tokens
120,427
Response Time (avg)
28.08s
Total Cost
$2.126
Rank
#34
Total Output Tokens
66,287
Response Time (avg)
18.87s
Total Cost
$1.318
Rank
#71
Total Output Tokens
102,691
Response Time (avg)
22.71s
Total Cost
$1.585
Recommended modelGPT-5.3-Codex (medium)
It has the best score here (9.1), while costing about 1.7x less than the other models in this comparison.
Detailed comparison
Metric
Claude Opus 4.6Claude Opus 4.6mediumArchived model: this model is no longer updated or tested on new tests.Release: 2026-02-05
6.3Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
6.4Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
9.1Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
8.4Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
Rank
#228
#212
#34
#71
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.…
10.0First-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.…
9.8First-attempt success score: 10.0 means no retryable target API or rate-limit failures before successful calls; tracked failures lower the score.…
Consistency
8.5Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
9.1Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
8.6Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
9.6Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
Attempts
66/69
66/69
69/69
69/69
Tests Correct
A test is fully passed only if every run passed for that test.Extra formatting: 4Wrong answer: 4Did not follow instructions: 1Response Time (avg)32.23sResponse Time (max)151.51sResponse Time (total)515.65sA 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: 4Extra formatting: 3Timed out: 1Response Time (avg)28.08sResponse Time (max)140.96sResponse Time (total)421.15sA 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: 4Did not follow instructions: 2Response Time (avg)18.87sResponse Time (max)100.93sResponse Time (total)433.94sA 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: 2Timed out: 1Response Time (avg)22.71sResponse Time (max)88.68sResponse Time (total)386.11sA test is fully passed only if every run passed for that test.…
Attempt pass rate
60.9%Attempt pass rate = passed attempts / total attempts across runs.…
62.3%Attempt pass rate = passed attempts / total attempts across runs.…
84.1%Attempt pass rate = passed attempts / total attempts across runs.…
89.9%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).…
1Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
4Flaky 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
66Total Runs…
66Total Runs…
69Total Runs…
69Total Runs…
Cost per result
22.777
15.182
7.753
7.924
Total Cost
$2.961
$2.126
$1.318
$1.585
Input Price
$5.000 / 1MInput Price…
$3.000 / 1MInput Price…
$1.750 / 1MInput Price…
$2.000 / 1MInput Price…
Output Price
$25.000 / 1MOutput Price…
$15.000 / 1MOutput Price…
$14.000 / 1MOutput Price…
$12.000 / 1MOutput Price…
Cache Read Price
$0.500 / 1MCache Read Price…
$0.300 / 1MCache Read Price…
$0.175 / 1MCache Read Price…
$0.200 / 1MCache Read Price…
Cache Write Price
$6.250 / 1MCache Write Price…
$3.750 / 1MCache Write Price…
N/ACache Write Price…
$0.375 / 1MCache Write Price…
Total Input Tokens
108,573Total Input Tokens…
106,316Total Input Tokens…
222,794Total Input Tokens…
176,208Total Input Tokens…
Output Tokens
69,994Output Tokens…
79,338Output Tokens…
7,357Output Tokens…
6,093Output Tokens…
Reasoning Tokens
26,728Reasoning Tokens…
41,089Reasoning Tokens…
58,930Reasoning Tokens…
96,598Reasoning Tokens…
Response Time (avg)
32.23sResponse Time (avg)…
28.08sResponse Time (avg)…
18.87sResponse Time (avg)…
22.71sResponse Time (avg)…
Response Time (max)
151.51sResponse Time (max)…
140.96sResponse Time (max)…
100.93sResponse Time (max)…
88.68sResponse Time (max)…
Response Time (total)
515.65sResponse Time (total)…
421.15sResponse Time (total)…
433.94sResponse Time (total)…
386.11sResponse Time (total)…
Parameters
~5T total (~500B active)
~1T total (~100B active)
~1.2T total (~80B active)
~1.2T total (~20B active)
Availability
Closed
Closed
Closed
Closed
Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.
Model generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#228 Claude Opus 4.6
medium
Reached the allocated time limit (300 seconds) without receiving showcase output.
Cost
$0.000
Time
300.0s
Tokens
0 tok
#212 Claude Sonnet 4.6
medium
Reached the allocated time limit (300 seconds) without receiving showcase output.
Cost
$0.000
Time
300.0s
Tokens
0 tok
#34 GPT-5.3-Codex
medium
Cost
$0.049
Time
54.9s
Tokens
3,580 tok
#71 Gemini 3.1 Pro Preview
medium
Cost
$0.115
Time
87.2s
Tokens
9,629 tok
Rank
-
Cost
-
Time
-
Tokens
-
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Category Breakdown
Agentic
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived 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.…
0msResponse Time (avg)…
0Total Input Tokens…
0Output Tokens…
0Reasoning Tokens…
Claude Sonnet 4.6Archived 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)61.91sResponse Time (max)61.91sResponse Time (total)61.91sA test is fully passed only if every run passed for that test.…
4.7Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
1.6Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
66.7%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.Timed out: 1Response Time (avg)56.17sResponse Time (max)56.17sResponse Time (total)56.17sA test is fully passed only if every run passed for that test.…
56.17sResponse Time (avg)…
83,912Total Input Tokens…
861Output Tokens…
4,592Reasoning Tokens…
Anti-AI Tricks
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived model: this model is no longer updated or tested on new tests.
6.4Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
5.8Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
66.7%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.Extra formatting: 2Response Time (avg)7.45sResponse Time (max)11.88sResponse Time (total)14.90sA test is fully passed only if every run passed for that test.…
7.45sResponse Time (avg)…
840Total Input Tokens…
986Output Tokens…
1,071Reasoning Tokens…
Claude Sonnet 4.6Archived model: this model is no longer updated or tested on new tests.
6.5Summarizes 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).…
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.Extra formatting: 1Wrong answer: 1Response Time (avg)2.98sResponse Time (max)4.95sResponse Time (total)5.97sA test is fully passed only if every run passed for that test.…
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)4.16sResponse Time (max)6.68sResponse Time (total)16.63sA 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)7.90sResponse Time (max)9.52sResponse Time (total)15.80sA test is fully passed only if every run passed for that test.…
7.90sResponse Time (avg)…
498Total Input Tokens…
112Output Tokens…
3,218Reasoning Tokens…
Coding
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived model: this model is no longer updated or tested on new tests.
5.7Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
7.1Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
44.4%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.Extra formatting: 1Did not follow instructions: 1Response Time (avg)30.10sResponse Time (max)35.63sResponse Time (total)90.31sA test is fully passed only if every run passed for that test.…
30.10sResponse Time (avg)…
8,522Total Input Tokens…
13,057Output Tokens…
4,121Reasoning Tokens…
Claude Sonnet 4.6Archived model: this model is no longer updated or tested on new tests.
5.7Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
6.6Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
44.4%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.Extra formatting: 1Wrong answer: 1Response Time (avg)33.29sResponse Time (max)35.76sResponse Time (total)99.86sA 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)19.50sResponse Time (max)27.96sResponse Time (total)58.49sA test is fully passed only if every run passed for that test.…
7.9Summarizes 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).…
66.7%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)40.17sResponse Time (max)88.68sResponse Time (total)120.52sA test is fully passed only if every run passed for that test.…
40.17sResponse Time (avg)…
8,124Total Input Tokens…
435Output Tokens…
41,247Reasoning Tokens…
Combined
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived 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)114.08sResponse Time (max)151.51sResponse Time (total)228.16sA test is fully passed only if every run passed for that test.…
114.08sResponse Time (avg)…
76,073Total Input Tokens…
33,018Output Tokens…
9,509Reasoning Tokens…
Claude Sonnet 4.6Archived 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)93.65sResponse Time (max)140.96sResponse Time (total)187.31sA 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)26.04sResponse Time (max)32.53sResponse Time (total)52.08sA 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)40.39sResponse Time (max)40.61sResponse Time (total)80.77sA test is fully passed only if every run passed for that test.…
40.39sResponse Time (avg)…
67,910Total Input Tokens…
3,687Output Tokens…
23,111Reasoning Tokens…
Data parsing and extraction
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived 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)7.37sResponse Time (max)7.37sResponse Time (total)7.37sA test is fully passed only if every run passed for that test.…
7.37sResponse Time (avg)…
8,676Total Input Tokens…
691Output Tokens…
757Reasoning Tokens…
Claude Sonnet 4.6Archived 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)13.90sResponse Time (max)13.90sResponse Time (total)13.90sA 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)3.07sResponse Time (max)3.59sResponse Time (total)6.15sA 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)…
7,265Total Input Tokens…
279Output Tokens…
3,904Reasoning Tokens…
Domain specific
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived 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: 2Extra formatting: 1Response Time (avg)42.53sResponse Time (max)83.40sResponse Time (total)85.06sA test is fully passed only if every run passed for that test.…
42.53sResponse Time (avg)…
632Total Input Tokens…
12,350Output Tokens…
7,100Reasoning Tokens…
Claude Sonnet 4.6Archived 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.Extra formatting: 1Timed out: 1Wrong answer: 1Response Time (avg)58.37sResponse Time (max)58.37sResponse Time (total)58.37sA test is fully passed only if every run passed for that test.…
5.9Summarizes 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).…
55.6%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: 2Response Time (avg)63.93sResponse Time (max)100.93sResponse Time (total)191.78sA test is fully passed only if every run passed for that test.…
7.7Summarizes 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).…
66.7%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)20.29sResponse Time (max)32.73sResponse Time (total)40.59sA test is fully passed only if every run passed for that test.…
20.29sResponse Time (avg)…
644Total Input Tokens…
18Output Tokens…
11,704Reasoning Tokens…
General Intelligence
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived 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)5.04sResponse Time (max)5.04sResponse Time (total)5.04sA test is fully passed only if every run passed for that test.…
5.04sResponse Time (avg)…
564Total Input Tokens…
188Output Tokens…
292Reasoning Tokens…
Claude Sonnet 4.6Archived 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)4.94sResponse Time (max)4.94sResponse Time (total)4.94sA test is fully passed only if every run passed for that test.…
4.6Summarizes 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)4.87sResponse Time (max)4.87sResponse Time (total)4.87sA 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)11.77sResponse Time (max)11.77sResponse Time (total)11.77sA test is fully passed only if every run passed for that test.…
11.77sResponse Time (avg)…
490Total Input Tokens…
108Output Tokens…
1,179Reasoning Tokens…
Instructions following
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived 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)2.43sResponse Time (max)2.43sResponse Time (total)2.43sA test is fully passed only if every run passed for that test.…
2.43sResponse Time (avg)…
792Total Input Tokens…
266Output Tokens…
467Reasoning Tokens…
Claude Sonnet 4.6Archived 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)2.61sResponse Time (max)2.61sResponse Time (total)2.61sA 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)3.04sResponse Time (max)3.44sResponse Time (total)6.07sA 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)9.56sResponse Time (max)9.56sResponse Time (total)9.56sA test is fully passed only if every run passed for that test.…
9.56sResponse Time (avg)…
621Total Input Tokens…
72Output Tokens…
2,236Reasoning Tokens…
Puzzle Solving
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived model: this model is no longer updated or tested on new tests.
7.7Summarizes 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).…
66.7%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)4.71sResponse Time (max)4.75sResponse Time (total)9.41sA test is fully passed only if every run passed for that test.…
4.71sResponse Time (avg)…
816Total Input Tokens…
532Output Tokens…
630Reasoning Tokens…
Claude Sonnet 4.6Archived 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)5.31sResponse Time (max)6.24sResponse Time (total)10.62sA test is fully passed only if every run passed for that test.…
9.0Summarizes 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).…
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.Did not follow instructions: 1Response Time (avg)5.05sResponse Time (max)8.73sResponse Time (total)15.15sA 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)6.90sResponse Time (max)8.49sResponse Time (total)13.79sA test is fully passed only if every run passed for that test.…
6.90sResponse Time (avg)…
570Total Input Tokens…
235Output Tokens…
3,128Reasoning Tokens…
Tool Calling
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived 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)9.73sResponse Time (max)9.73sResponse Time (total)9.73sA test is fully passed only if every run passed for that test.…
9.73sResponse Time (avg)…
11,454Total Input Tokens…
861Output Tokens…
329Reasoning Tokens…
Claude Sonnet 4.6Archived 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)7.48sResponse Time (max)7.48sResponse Time (total)7.48sA 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)6.37sResponse Time (max)6.37sResponse Time (total)6.37sA 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)23.15sResponse Time (max)23.15sResponse Time (total)23.15sA test is fully passed only if every run passed for that test.…
23.15sResponse Time (avg)…
6,018Total Input Tokens…
274Output Tokens…
982Reasoning Tokens…
Trivia
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Claude Opus 4.6Archived 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)63.24sResponse Time (max)63.24sResponse Time (total)63.24sA test is fully passed only if every run passed for that test.…
63.24sResponse Time (avg)…
204Total Input Tokens…
8,045Output Tokens…
2,452Reasoning Tokens…
Claude Sonnet 4.6Archived 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)30.09sResponse Time (max)30.09sResponse Time (total)30.09sA test is fully passed only if every run passed for that test.…
2.8Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
1.6Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
33.3%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)14.43sResponse Time (max)14.43sResponse Time (total)14.43sA 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)6.27sResponse Time (max)6.27sResponse Time (total)6.27sA test is fully passed only if every run passed for that test.…