Loading analysis results...
Please wait while we prepare the detailed comparison.
Please wait while we prepare the detailed comparison.
| 🏆 Best Hybrid Score | grok-4-0709 | 86.9% |
| 📉 Worst Hybrid Score | openai/gpt-4.1-nano | 61.3% |
| 🤔 Most Differentiating Prompt | User: How does the age of a record affect its exemption from disclosure under Section 8 of the RTI Act? | σ = 0.24 |
| 👯 Most Semantically Similar Pair | gpt-4o-2024-08-06 vs openai/gpt-4o | 98.4% |
A consolidated overview of performance and semantic consistency metrics.
Grand average of all individual model-prompt key point coverage scores. StdDev (±) reflects variability around this grand mean, also in percentage points. A smaller StdDev suggests more consistent coverage scores across all model-prompt pairs; a larger StdDev indicates more diverse performance.
Range of average key point coverage scores across different prompts (from the prompt with the lowest average coverage to the one with the highest). A large spread indicates substantial differences in how challenging prompts were or how models performed on them.
Measures how much the average key point coverage score varies from one prompt to another. A high value (e.g., >20-25%) suggests that average performance was quite different across prompts; a low value suggests more consistent average performance from prompt to prompt.
Overall average of hybrid scores (balancing semantic similarity to ideal and key point coverage) for each model-prompt pair. Higher is generally better. A smaller StdDev suggests more consistent hybrid performance across all model-prompt pairs.
Models: openai:gpt-4o-2024-05-13, openai:gpt-4o-2024-08-06, openai:gpt-4o-2024-11-20, openrouter:anthropic/claude-3.5-haiku, openrouter:anthropic/claude-sonnet-4, openrouter:cohere/command-a, openrouter:deepseek/deepseek-chat-v3-0324, openrouter:deepseek/deepseek-r1, openrouter:google/gemini-2.5-flash-preview-05-20, openrouter:google/gemini-2.5-pro-preview-05-06, openrouter:mistralai/mistral-large-2411, openrouter:mistralai/mistral-medium-3, openrouter:openai/gpt-4.1-mini, openrouter:openai/gpt-4.1-nano, openrouter:openai/gpt-4.1, openrouter:openai/gpt-4o-mini, openrouter:openai/gpt-4o, openrouter:openai/o4-mini, openrouter:x-ai/grok-3-mini-beta, openrouter:x-ai/grok-3, xai:grok-4-0709
E.g. rti-exemptions: User: What are some key categories of information that are exempt from disclosure under Section 8 of...
Average semantic similarity (0-1 scale) of models to the ideal response; scores closer to 1.0 are better. The StdDev shows how consistently models achieve this. A very low StdDev (e.g., <0.05) often means models performed very similarly on this metric.
| Metric | Value | Explanation |
|---|---|---|
| Overall Average Key Point Coverage | 75.4% (±25.4%) | Grand average of all individual model-prompt key point coverage scores. StdDev (±) reflects variability around this grand mean, also in percentage points. A smaller StdDev suggests more consistent coverage scores across all model-prompt pairs; a larger StdDev indicates more diverse performance. |
| Avg. Prompt Coverage Range | 45% - 100% (Spread: 55 pp) | Range of average key point coverage scores across different prompts (from the prompt with the lowest average coverage to the one with the highest). A large spread indicates substantial differences in how challenging prompts were or how models performed on them. |
| StdDev of Avg. Prompt Coverage | 17.3% | Measures how much the average key point coverage score varies from one prompt to another. A high value (e.g., >20-25%) suggests that average performance was quite different across prompts; a low value suggests more consistent average performance from prompt to prompt. |
| Overall Average Hybrid Score | 75.5% (±16.1%) | Overall average of hybrid scores (balancing semantic similarity to ideal and key point coverage) for each model-prompt pair. Higher is generally better. A smaller StdDev suggests more consistent hybrid performance across all model-prompt pairs. |
| Number of Models Evaluated | 21 | Models: openai:gpt-4o-2024-05-13, openai:gpt-4o-2024-08-06, openai:gpt-4o-2024-11-20, openrouter:anthropic/claude-3.5-haiku, openrouter:anthropic/claude-sonnet-4, openrouter:cohere/command-a, openrouter:deepseek/deepseek-chat-v3-0324, openrouter:deepseek/deepseek-r1, openrouter:google/gemini-2.5-flash-preview-05-20, openrouter:google/gemini-2.5-pro-preview-05-06, openrouter:mistralai/mistral-large-2411, openrouter:mistralai/mistral-medium-3, openrouter:openai/gpt-4.1-mini, openrouter:openai/gpt-4.1-nano, openrouter:openai/gpt-4.1, openrouter:openai/gpt-4o-mini, openrouter:openai/gpt-4o, openrouter:openai/o4-mini, openrouter:x-ai/grok-3-mini-beta, openrouter:x-ai/grok-3, xai:grok-4-0709 |
| Number of Prompts Analyzed | 13 | E.g. rti-exemptions: User: What are some key categories of information that are exempt from disclosure under Section 8 of... |
| Average Semantic Similarity to Ideal | 0.759 (±0.024) | Average semantic similarity (0-1 scale) of models to the ideal response; scores closer to 1.0 are better. The StdDev shows how consistently models achieve this. A very low StdDev (e.g., <0.05) often means models performed very similarly on this metric. |
Average key point coverage extent for each model across all prompts.
| Prompts vs. Models | GPT 4o 2024 05 13 | GPT 4o 2024 08 06 | GPT 4o 2024 11 20 | Claude 3.5 Haiku | Claude Sonnet 4 | Command A | Deepseek Chat V3 0324 | Deepseek R1 | Gemini 2.5 Flash Preview 05 20 | Gemini 2.5 Pro Preview 05 06 | Mistral Large 2411 | Mistral Medium 3 | GPT 4.1 Mini | GPT 4.1 Nano | GPT 4.1 | GPT 4o Mini | GPT 4o | O4 Mini | Grok 3 Mini Beta | Grok 3 | Grok 4 0709 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Score | 11th 76.9% | 9th 78.7% | 5th 81.1% | 19th 64.8% | 10th 77.8% | 17th 70.0% | 12th 75.3% | 8th 79.8% | 2nd 88.2% | 18th 66.8% | 13th 74.2% | 13th 74.2% | 15th 71.1% | 21st 52.4% | 4th 81.4% | 20th 58.5% | 7th 80.0% | 16th 70.9% | 6th 80.9% | 3rd 84.8% | 1st 94.7% | |
| 47.3% | 25% | 59% | 63% | 13% | 32% | 38% | 25% | 47% | 100% | 50% | 94% | 53% | 38% | 0% | 53% | 13% | 53% | 19% | 56% | 63% | 100% | |
| 60.7% | 63% | 63% | 63% | 63% | 75% | 56% | 63% | 69% | 25% | 63% | 63% | 50% | 63% | 50% | 75% | 63% | 69% | 50% | 63% | 63% | 63% | |
| 86.1% | 85% | 85% | 83% | 80% | 90% | 98% | 88% | 88% | 93% | 75% | 85% | 85% | 90% | 70% | 83% | 65% | 88% | 98% | 95% | 90% | 95% | |
| 82.3% | 81% | 98% | 100% | 71% | 94% | 85% | 96% | 100% | 100% | 33% | 90% | 67% | 60% | 35% | 83% | 69% | 100% | 69% | 100% | 100% | 98% | |
| 64.1% | 75% | 63% | 75% | 75% | 53% | 56% | 72% | 100% | 75% | 50% | 50% | 34% | 75% | 31% | 78% | 53% | 72% | 50% | 59% | 75% | 75% | |
| 100.0% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | |
| 83.4% | 80% | 90% | 98% | 55% | 95% | 90% | 90% | 98% | 100% | 58% | 68% | 90% | 83% | 55% | 90% | 55% | 88% | 98% | 70% | 100% | 100% | |
| 60.4% | 63% | 54% | 46% | 75% | 63% | 54% | 42% | 71% | 54% | 100% | 83% | 46% | 38% | 50% | 46% | 38% | 54% | 75% | 75% | 42% | 100% | |
| 92.1% | 97% | 94% | 94% | 100% | 94% | 94% | 97% | 91% | 100% | 53% | 88% | 100% | 84% | 78% | 100% | 91% | 91% | 88% | 100% | 100% | 100% | |
| 83.9% | 100% | 67% | 100% | 63% | 100% | 67% | 100% | 67% | 100% | 33% | 63% | 67% | 67% | 100% | 100% | 67% | 100% | 100% | 100% | 100% | 100% | |
| 77.3% | 75% | 100% | 100% | 34% | 78% | 84% | 81% | 100% | 100% | 53% | 100% | 72% | 88% | 34% | 75% | 47% | 75% | 75% | 53% | 100% | 100% | |
| 96.5% | 100% | 100% | 100% | 100% | 100% | 88% | 100% | 100% | 100% | 100% | 67% | 100% | 100% | 71% | 100% | 100% | 100% | 100% | 100% | 100% | 100% | |
| 45.4% | 56% | 50% | 32% | 13% | 38% | 0% | 25% | 7% | 100% | 100% | 13% | 100% | 38% | 7% | 75% | 0% | 50% | 0% | 81% | 69% | 100% |