Solar Decide (default) leads on average score with 1.0 vs 0.8. Nimble v3 9B (default) has the lower benchmark cost at ~$0.001 vs $0.009. Nimble v3 9B (default) is faster at 363ms vs 6.01s, with pass rates of 4.4% vs 14.5%.
Last updated at:
2026-10-16
Compared models
Candidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
Rank
#430
Total Output Tokens
0
Response Time (avg)
363ms
Total Cost
~$0.001Estimated GPU electricity cost only. It excludes the rest of the computer and hardware ownership. NVIDIA GeForce RTX 3090: 0.0021 h × 0.377 kW × €0.2896/kWh × 1.138 USD/EUR = ~$0.000263.
Rank
#428
Total Output Tokens
51
Response Time (avg)
6.01s
Total Cost
$0.009
Recommended modelSolar Decide (default)
It has the strongest score in this comparison (1.0) and the best overall balance of cost and response time across 2 models.
Detailed comparison
Metric
Nimble v3 9BNimble v3 9BdefaultCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.21/69 attempts. Choices supplied by adapters. Supports 7/23 tests. Other tests are unsupported, not failed. Score uses the full suite.Release: 2026-10-16
Solar DecideSolar Decidedefault21/69 attempts. Choices supplied by adapters. Supports 7/23 tests. Other tests are unsupported, not failed. Score uses the full suite.Release: 2026-09-28
Metric
Nimble v3 9BNimble v3 9BdefaultCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.21/69 attempts. Choices supplied by adapters. Supports 7/23 tests. Other tests are unsupported, not failed. Score uses the full suite.Release: 2026-10-16
Solar DecideSolar Decidedefault21/69 attempts. Choices supplied by adapters. Supports 7/23 tests. Other tests are unsupported, not failed. Score uses the full suite.Release: 2026-09-28
Score
0.8Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
1.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
Rank
#430
#428
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
3.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
2.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
Attempts
21/69
21/69
Tests Correct
A test is fully passed only if every run passed for that test.Wrong answer: 5Did not follow instructions: 1Response Time (avg)363msResponse Time (max)1.24sResponse Time (total)2.54sA 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: 5Response Time (avg)6.01sResponse Time (max)13.10sResponse Time (total)42.05sA test is fully passed only if every run passed for that test.…
Attempt pass rate
4.4%Attempt pass rate = passed attempts / total attempts across runs.…
14.5%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).…
3Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
Total Runs
21Total Runs…
21Total Runs…
Cost per result
~0.027Estimated GPU electricity cost only. It excludes the rest of the computer and hardware ownership. NVIDIA GeForce RTX 3090: 0.0021 h × 0.377 kW × €0.2896/kWh × 1.138 USD/EUR = ~$0.000263.
0.119
Total Cost
~$0.001Estimated GPU electricity cost only. It excludes the rest of the computer and hardware ownership. NVIDIA GeForce RTX 3090: 0.0021 h × 0.377 kW × €0.2896/kWh × 1.138 USD/EUR = ~$0.000263.
$0.009
Input Price
N/AInput Price…
$0.050 / 1MInput Price…
Output Price
N/AOutput Price…
$0.000 / 1MOutput Price…
Cache Read Price
N/ACache Read Price…
$0.050 / 1MCache Read Price…
Cache Write Price
N/ACache Write Price…
N/ACache Write Price…
Total Input Tokens
103,770Total Input Tokens…
47,586Total Input Tokens…
Output Tokens
0Output Tokens…
51Output Tokens…
Reasoning Tokens
0Reasoning Tokens…
0Reasoning Tokens…
Response Time (avg)
363msResponse Time (avg)…
6.01sResponse Time (avg)…
Response Time (max)
1.24sResponse Time (max)…
13.10sResponse Time (max)…
Response Time (total)
2.54sResponse Time (total)…
42.05sResponse Time (total)…
Parameters
9B
~35B total (~3B active)
Availability
Weights available
Closed
Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.
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
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
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.…
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…
Anti-AI Tricks
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
2.5Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
2.5Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
25.0%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)73msResponse Time (max)73msResponse Time (total)73msA test is fully passed only if every run passed for that test.…
0.8Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
2.5Consistency 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.Wrong answer: 1Response Time (avg)2.41sResponse Time (max)2.41sResponse Time (total)2.41sA test is fully passed only if every run passed for that test.…
2.41sResponse Time (avg)…
1,239Total Input Tokens…
3Output Tokens…
0Reasoning Tokens…
Coding
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
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.…
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…
Combined
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
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.…
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…
Data parsing and extraction
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
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: 2Response Time (avg)880msResponse Time (max)1.24sResponse Time (total)1.76sA 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.…
1.7Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
50.0%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: 2Response Time (avg)5.00sResponse Time (max)6.86sResponse Time (total)10.00sA test is fully passed only if every run passed for that test.…
5.00sResponse Time (avg)…
31,275Total Input Tokens…
18Output Tokens…
0Reasoning Tokens…
Domain specific
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
2.0Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
6.7Consistency 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: 2Response Time (avg)80msResponse Time (max)91msResponse Time (total)160msA test is fully passed only if every run passed for that test.…
4.3Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
3.9Consistency 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.Wrong answer: 1Response Time (avg)7.26sResponse Time (max)13.10sResponse Time (total)14.52sA test is fully passed only if every run passed for that test.…
7.26sResponse Time (avg)…
2,580Total Input Tokens…
6Output Tokens…
0Reasoning Tokens…
General Intelligence
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
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.…
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…
Instructions following
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
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.Wrong answer: 1Response Time (avg)469msResponse Time (max)469msResponse Time (total)469msA 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)11.66sResponse Time (max)11.66sResponse Time (total)11.66sA test is fully passed only if every run passed for that test.…
11.66sResponse Time (avg)…
11,196Total Input Tokens…
21Output Tokens…
0Reasoning Tokens…
Puzzle Solving
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
1.7Summarizes 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.Did not follow instructions: 1Response Time (avg)79msResponse Time (max)79msResponse Time (total)79msA test is fully passed only if every run passed for that test.…
1.0Summarizes 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.Wrong answer: 1Response Time (avg)3.47sResponse Time (max)3.47sResponse Time (total)3.47sA test is fully passed only if every run passed for that test.…
3.47sResponse Time (avg)…
1,296Total Input Tokens…
3Output Tokens…
0Reasoning Tokens…
Tool Calling
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
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.…
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…
Trivia
Score
Consistency
Attempt pass rate
Flaky tests
Tests Correct
Response Time (avg)
Input Tokens
Output Tokens
Reasoning Tokens
Nimble v3 9BCandidate responses were generated on local hardware. OpenRouter was used only for judging. nimble-v3:9b-q8_0 · Q8_0 · Ollama 0.40.5 · NVIDIA GeForce RTX 3090.
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.…
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.…