Qwen3.8 27B (low) leads on average score with 7.9 vs 7.7. Qwen3.8 27B (low) has the lower benchmark cost at ~$0.071 vs $0.228. GLM 5 (medium) is faster at 33.54s vs 39.11s, with pass rates of 68.2% vs 78.8%.
Last updated at:
2026-08-20
Candidate responses were generated on local hardware. OpenRouter was used only for judging.
Rank
#61
Total Output Tokens
169,329
Response Time (avg)
39.11s
Total Cost
~$0.071Estimated GPU electricity cost only. It excludes the rest of the computer and hardware ownership. NVIDIA GeForce RTX 3090: 0.7171 h × 0.300 kW × €0.2896/kWh × 1.138 USD/EUR = ~$0.070898.
Rank
#72
Total Output Tokens
124,566
Response Time (avg)
33.54s
Total Cost
$0.228
Recommended modelQwen3.8 27B (low)
It has the best score here (7.9), while costing about 3.2x less than GLM 5 (medium).
Detailed comparison
Metric
Qwen3.8 27BQwen3.8 27BlowCandidate responses were generated on local hardware. OpenRouter was used only for judging.Release: 2026-08-14
GLM 5GLM 5medium63/66 attempts (Target: 3 repeats per test)Release: 2026-02-12
Metric
Qwen3.8 27BQwen3.8 27BlowCandidate responses were generated on local hardware. OpenRouter was used only for judging.Release: 2026-08-14
GLM 5GLM 5medium63/66 attempts (Target: 3 repeats per test)Release: 2026-02-12
Score
7.9Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
7.7Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
Rank
#61
#72
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.…
Consistency
10.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
8.1Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
Attempts
66/66
63/66
Tests Correct
A test is fully passed only if every run passed for that test.Wrong answer: 4No answer: 2Did not follow instructions: 1Response Time (avg)39.11sResponse Time (max)376.04sResponse Time (total)860.51sA test is fully passed only if every run passed for that test.…
68.2%Attempt pass rate = passed attempts / total attempts across runs.…
78.8%Attempt pass rate = passed attempts / total attempts across runs.…
Flaky tests
0Flaky 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).…
Total Runs
66Total Runs…
63Total Runs…
Cost per result
~0.473Estimated GPU electricity cost only. It excludes the rest of the computer and hardware ownership. NVIDIA GeForce RTX 3090: 0.7171 h × 0.300 kW × €0.2896/kWh × 1.138 USD/EUR = ~$0.070898.
1.668
Total Cost
~$0.071Estimated GPU electricity cost only. It excludes the rest of the computer and hardware ownership. NVIDIA GeForce RTX 3090: 0.7171 h × 0.300 kW × €0.2896/kWh × 1.138 USD/EUR = ~$0.070898.
$0.228
Input Price
N/AInput Price…
$0.600 / 1MInput Price…
Output Price
N/AOutput Price…
$1.920 / 1MOutput Price…
Total Input Tokens
99,705Total Input Tokens…
35,224Total Input Tokens…
Output Tokens
1,545Output Tokens…
21,570Output Tokens…
Reasoning Tokens
167,784Reasoning Tokens…
102,996Reasoning Tokens…
Response Time (avg)
39.11sResponse Time (avg)…
33.54sResponse Time (avg)…
Response Time (max)
376.04sResponse Time (max)…
99.85sResponse Time (max)…
Response Time (total)
860.51sResponse Time (total)…
435.99sResponse Time (total)…
Parameters
27.3B
744B total (40B active)
Availability
Weights available
Open source
Model generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#61 Qwen3.8 27B
lowCandidate responses were generated on local hardware. OpenRouter was used only for judging.
Cost
~$0.003Estimated GPU electricity cost only. It excludes the rest of the computer and hardware ownership. NVIDIA GeForce RTX 3090: 0.0258 h × 0.300 kW × €0.2896/kWh × 1.138 USD/EUR = ~$0.000000.
Time
92.8s
Tokens
6,902 tok
#72 GLM 5
medium
Cost
$0.005
Time
20.7s
Tokens
2,068 tok
Score
-
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
Anti-AI Tricks
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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)3.02sResponse Time (max)5.68sResponse Time (total)12.07sA 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)23.66sResponse Time (max)25.06sResponse Time (total)47.32sA test is fully passed only if every run passed for that test.…
23.66sResponse Time (avg)…
555Total Input Tokens…
480Output Tokens…
7,056Reasoning Tokens…
Coding
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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.No answer: 1Response Time (avg)140.92sResponse Time (max)376.04sResponse Time (total)422.75sA 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)74.30sResponse Time (max)99.85sResponse Time (total)222.91sA test is fully passed only if every run passed for that test.…
74.30sResponse Time (avg)…
7,254Total Input Tokens…
2,997Output Tokens…
52,930Reasoning Tokens…
Combined
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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)90.63sResponse Time (max)143.71sResponse Time (total)181.27sA test is fully passed only if every run passed for that test.…
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)28.96sResponse Time (max)28.96sResponse Time (total)28.96sA test is fully passed only if every run passed for that test.…
28.96sResponse Time (avg)…
12,804Total Input Tokens…
662Output Tokens…
3,242Reasoning Tokens…
Data parsing and extraction
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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)8.03sResponse Time (max)9.09sResponse Time (total)16.06sA test is fully passed only if every run passed for that test.…
7.1Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
5.6Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
83.3%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.No answer: 1Response Time (avg)8.90sResponse Time (max)8.90sResponse Time (total)8.90sA test is fully passed only if every run passed for that test.…
8.90sResponse Time (avg)…
5,508Total Input Tokens…
567Output Tokens…
3,734Reasoning Tokens…
Domain specific
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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)60.78sResponse Time (max)120.94sResponse Time (total)182.34sA 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: 2Timed out: 1Response Time (avg)0msResponse Time (max)0msResponse Time (total)0msA test is fully passed only if every run passed for that test.…
0msResponse Time (avg)…
260Total Input Tokens…
13,176Output Tokens…
14,137Reasoning Tokens…
General Intelligence
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
5.0Summarizes 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)5.37sResponse Time (max)5.37sResponse Time (total)5.37sA test is fully passed only if every run passed for that test.…
6.1Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
3.1Consistency 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.Did not follow instructions: 1Response Time (avg)14.69sResponse Time (max)14.69sResponse Time (total)14.69sA test is fully passed only if every run passed for that test.…
14.69sResponse Time (avg)…
477Total Input Tokens…
2,020Output Tokens…
2,248Reasoning Tokens…
Instructions following
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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)2.87sResponse Time (total)4.86sA 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)7.25sResponse Time (max)7.25sResponse Time (total)7.25sA test is fully passed only if every run passed for that test.…
7.25sResponse Time (avg)…
636Total Input Tokens…
1,001Output Tokens…
2,129Reasoning Tokens…
Puzzle Solving
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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)4.91sResponse Time (max)8.81sResponse Time (total)14.74sA 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)11.33sResponse Time (max)16.34sResponse Time (total)22.66sA test is fully passed only if every run passed for that test.…
11.33sResponse Time (avg)…
609Total Input Tokens…
33Output Tokens…
4,076Reasoning Tokens…
Tool Calling
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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)8.42sResponse Time (max)8.42sResponse Time (total)8.42sA 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)15.93sResponse Time (max)15.93sResponse Time (total)15.93sA test is fully passed only if every run passed for that test.…
15.93sResponse Time (avg)…
6,935Total Input Tokens…
233Output Tokens…
994Reasoning Tokens…
Trivia
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Qwen3.8 27BCandidate responses were generated on local hardware. OpenRouter was used only for judging.
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.No answer: 1Response Time (avg)12.64sResponse Time (max)12.64sResponse Time (total)12.64sA 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)67.37sResponse Time (max)67.37sResponse Time (total)67.37sA test is fully passed only if every run passed for that test.…