Run Your Own LLM Evaluation
Test any models with your prompts. Start free, no credit card required.
Coding Performance with 10 Evaluators — Run
Comprehensive evaluation of 2 language models across 1 system prompt with rigorous benchmarking and scoring criteria.
5.14
glm-5.3
5.00
Spread: 0.28 pts
1676ms
Response time
80
8 total responses
Executive Insights
Key takeaways from this evaluation
Top Performer
glm-5.3
5.14
0.28 pts ahead of #2
Model Rankings
Ranked by overall performance score
glm-5.3
z-ai/glm-5.3
Derived from how judges ranked this response against the others — not an absolute quality rating.
Responses
4
Avg Latency
473ms
Cost
$0.0205
claude-fable-5.1
anthropic/claude-fable-5.1
Derived from how judges ranked this response against the others — not an absolute quality rating.
Responses
4
Avg Latency
2879ms
Cost
$0.1585
Evaluator Consensus
How 9 evaluator models ranked the candidates via blind comparison
majority Agreement
6 of 9 evaluators agree on the top model
claude-fable-5.1
Avg Rank
1.3
Range
#1–2
#1 Votes
6/9
Latency
2879ms
glm-5.3
Avg Rank
1.7
Range
#1–2
#1 Votes
3/9
Latency
473ms
gpt-5.4-mini
gemini-3.1-flash-lite-preview
claude-sonnet-4.6
minimax-m2.7
kimi-k2.5
deepseek-v3.2
mistral-small-2603
qwen3.5-27b
nova-2-lite-v1
Score Comparison
Visual comparison of all model scores
Performance by System Prompt
How each model performs across different evaluation contexts
Top Performer
glm-5.3
5.14
Performance by Test Prompt
Model results broken down by individual test prompts
| Test Prompt | Avg Score |
|---|---|
Javascript Function 2 responses | 5.00 |
Write an Interval Merge Function 2 responses | 5.00 |
Debug Python 2 responses | 5.00 |
Refactor Javascript 2 responses | 5.00 |
About This Evaluation
Methodology, criteria weights, and evaluation confidence
8
Total Responses
80
Total Evaluations
