Grok 4.20 leads on average score with 3.4 vs 0.8. Nimble v3 9B (default) has the lower benchmark cost at ~$0.001 vs $0.057. Nimble v3 9B (default) is faster at 363ms vs 1.11s, with pass rates of 4.4% vs 26.1%.
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
#401
Total Output Tokens
1,923
Response Time (avg)
1.11s
Total Cost
$0.057
Recommended modelGrok 4.20
It has the strongest score in this comparison (3.4) 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
Grok 4.20Grok 4.20none54/69 attempts (Target: 3 repeats per test)Release: 2026-03-31
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
Grok 4.20Grok 4.20none54/69 attempts (Target: 3 repeats per test)Release: 2026-03-31
Score
0.8Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
3.4Summarizes broad quality across our full private benchmark suite, so ranking reflects consistent performance.…
Rank
#430
#401
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
3.0Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
7.8Consistency score reflects run-to-run stability (10 = very consistent, even if consistently wrong).…
Attempts
21/69
54/69
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: 10Extra formatting: 1Invalid tool call: 1Response Time (avg)1.11sResponse Time (max)6.04sResponse Time (total)19.96sA 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.…
26.1%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).…
0Flaky tests had mixed outcomes across runs (at least one pass and one fail).…
Total Runs
21Total Runs…
54Total 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.
1.570
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.
$0.057
Input Price
N/AInput Price…
$1.250 / 1MInput Price…
Output Price
N/AOutput Price…
$2.500 / 1MOutput Price…
Cache Read Price
N/ACache Read Price…
$0.200 / 1MCache Read Price…
Cache Write Price
N/ACache Write Price…
N/ACache Write Price…
Total Input Tokens
103,770Total Input Tokens…
41,313Total Input Tokens…
Output Tokens
0Output Tokens…
1,923Output Tokens…
Reasoning Tokens
0Reasoning Tokens…
0Reasoning Tokens…
Response Time (avg)
363msResponse Time (avg)…
1.11sResponse Time (avg)…
Response Time (max)
1.24sResponse Time (max)…
6.04sResponse Time (max)…
Response Time (total)
2.54sResponse Time (total)…
19.96sResponse Time (total)…
Parameters
9B
~1T total (~100B 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.
Model generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#430 Nimble v3 9B
defaultCandidate 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.
No showcase result has been generated for this model yet.
Cost
N/A
Time
-
Tokens
0 tok
#401 xAI: Grok 4.20
none
Cost
$0.004
Time
6.5s
Tokens
1,367 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
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.…
4.8Summarizes 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).…
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.Wrong answer: 3Response Time (avg)501msResponse Time (max)839msResponse Time (total)2.01sA test is fully passed only if every run passed for that test.…
501msResponse Time (avg)…
1,986Total Input Tokens…
267Output 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.…
1.1Summarizes 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).…
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)1.22sResponse Time (max)1.22sResponse Time (total)1.22sA test is fully passed only if every run passed for that test.…
1.22sResponse Time (avg)…
1,074Total Input Tokens…
312Output 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.…
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.Invalid tool call: 1Response Time (avg)6.04sResponse Time (max)6.04sResponse Time (total)6.04sA test is fully passed only if every run passed for that test.…
6.04sResponse Time (avg)…
17,673Total Input Tokens…
282Output 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.…
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)522msResponse Time (max)537msResponse Time (total)1.04sA test is fully passed only if every run passed for that test.…
522msResponse Time (avg)…
7,749Total Input Tokens…
207Output 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.…
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)687msResponse Time (max)821msResponse Time (total)2.06sA test is fully passed only if every run passed for that test.…
687msResponse Time (avg)…
1,746Total Input Tokens…
325Output 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.…
4.8Summarizes 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).…
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)659msResponse Time (max)659msResponse Time (total)659msA test is fully passed only if every run passed for that test.…
659msResponse Time (avg)…
819Total Input Tokens…
83Output 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.…
6.3Summarizes 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).…
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.Wrong answer: 1Response Time (avg)445msResponse Time (max)505msResponse Time (total)889msA test is fully passed only if every run passed for that test.…
445msResponse Time (avg)…
1,350Total Input Tokens…
60Output 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.…
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)473msResponse Time (max)502msResponse Time (total)1.42sA test is fully passed only if every run passed for that test.…
473msResponse Time (avg)…
1,671Total Input Tokens…
198Output 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.…
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)4.63sResponse Time (max)4.63sResponse Time (total)4.63sA test is fully passed only if every run passed for that test.…
4.63sResponse Time (avg)…
7,245Total Input Tokens…
189Output 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.…