Text Replacement
Tests deterministic text transformations: renaming characters/locations, expanding contractions, tense rewriting, POV shifts, gender swaps, combined transformations, and word avoidance. Scored by checking each expected change independently.
Tense rewriting: past to present
Text Editing
Performance Score Distribution (Top 20)
Click a model name to view its detail page.
Price-Performance Score Distribution (Top 20)
Click a model name to view its detail page.
| Score | Cost | Time | ||
|---|---|---|---|---|
| Gemini 2.5 Flash Lite | 98% | $0.0003 | 1.5s | |
| Mistral Small 3.2 24B | 99% | $0.0002 | 5.9s | |
| Ministral 3 14B | 99% | $0.0002 | 3.8s | |
| Mistral NeMO | 98% | $0.0002 | 3.0s | |
| Ministral 3 8B | 98% | $0.0002 | 3.7s | |
| Ministral 8B | 97% | $0.0001 | 3.6s | |
| Qwen 2.5 72B | 99% | $0.0003 | 10.4s | |
| Mistral Large 3 | 99% | $0.0010 | 7.4s | |
| Qwen 3.5 Plus (2026-02-15) | 99% | $0.0015 | 6.6s | |
| Rocinante 12B | 69% | $0.0004 | 8.1s | |
| Mistral Small Creative | 96% | $0.0002 | 2.9s | |
| Arcee AI: Trinity Large (Preview) | 99% | $0.0000 | 21.2s | |
| Llama 3.1 8B | 96% | $0.0001 | 11.5s | |
| Writer: Palmyra X5 | 100% | $0.0033 | 9.2s | |
| Gemini 3 Flash (Preview) | 96% | $0.0018 | 3.2s | |
| Mistral Medium 3.1 | 96% | $0.0012 | 6.5s | |
| Hermes 3 70B | 95% | $0.0003 | 13.9s | |
| Llama 3.1 70B | 96% | $0.0007 | 26.0s | |
| Mistral Large 2 | 99% | $0.0041 | 7.2s | |
| Mistral Large | 99% | $0.0041 | 7.1s | |
Most Stable Models (Top 20)
Ranked by stability (median × consistency). Click a model name to view its detail page.
| Score | Consistency | Stability | ||
|---|---|---|---|---|
| Claude Sonnet 4.5 | 100% | 100% | 100% | |
| Claude Opus 4.6 (Reasoning) | 100% | 99% | 99% | |
| Claude Sonnet 4 | 100% | 99% | 99% | |
| Claude Opus 4.6 | 99% | 100% | 99% | |
| Gemini 3 Pro (Preview) | 99% | 100% | 99% | |
| Mistral Large 3 | 99% | 100% | 99% | |
| Claude 3.7 Sonnet | 99% | 100% | 99% | |
| Mistral Large 2 | 99% | 100% | 99% | |
| Mistral Large | 99% | 100% | 99% | |
| Mistral Small 3.2 24B | 99% | 100% | 99% | |
| Arcee AI: Trinity Large (Preview) | 99% | 100% | 99% | |
| Claude Opus 4.5 | 99% | 100% | 99% | |
| Qwen 3.5 Plus (2026-02-15) | 99% | 100% | 99% | |
| Writer: Palmyra X5 | 100% | 99% | 99% | |
| Claude 3.5 Sonnet | 99% | 99% | 99% | |
| Grok 4 | 100% | 99% | 99% | |
| Gemini 3.1 Pro (Preview) | 99% | 100% | 99% | |
| Ministral 3 14B | 99% | 100% | 99% | |
| Ministral 3 8B | 98% | 100% | 98% | |
| Claude Sonnet 4.6 (Reasoning) | 99% | 97% | 97% | |
Top Overall Models (Top 20)
Ranked by composite score (performance, cost, speed & stability). Click a model name to view its detail page.
| Score | Cost | Speed | Stability | ||
|---|---|---|---|---|---|
| Mistral Small 3.2 24B | 99% | $0.0002 | 5.9s | 99% | |
| Ministral 3 14B | 99% | $0.0002 | 3.8s | 99% | |
| Ministral 3 8B | 98% | $0.0002 | 3.7s | 98% | |
| Mistral Large 3 | 99% | $0.0010 | 7.4s | 99% | |
| Qwen 3.5 Plus (2026-02-15) | 99% | $0.0015 | 6.6s | 99% | |
| Mistral NeMO | 98% | $0.0002 | 3.0s | 97% | |
| Gemini 2.5 Flash Lite | 98% | $0.0003 | 1.5s | 96% | |
| Arcee AI: Trinity Large (Preview) | 99% | $0.0000 | 21.2s | 99% | |
| Qwen 2.5 72B | 99% | $0.0003 | 10.4s | 96% | |
| Writer: Palmyra X5 | 100% | $0.0033 | 9.2s | 99% | |
| Mistral Large | 99% | $0.0041 | 7.1s | 99% | |
| Mistral Large 2 | 99% | $0.0041 | 7.2s | 99% | |
| Ministral 8B | 97% | $0.0001 | 3.6s | 93% | |
| Gemini 3 Flash (Preview) | 96% | $0.0018 | 3.2s | 95% | |
| Mistral Medium 3.1 | 96% | $0.0012 | 6.5s | 95% | |
| Llama 3.1 Nemotron 70B | 97% | $0.0013 | 13.9s | 97% | |
| Claude Haiku 4.5 | 97% | $0.0033 | 3.3s | 96% | |
| Mistral Small Creative | 96% | $0.0002 | 2.9s | 91% | |
| Llama 3.1 8B | 96% | $0.0001 | 11.5s | 91% | |
| GPT-4o, Aug. 6th (temp=0) | 97% | $0.0063 | 2.6s | 97% | |
| Median | Evaluator | Top 3 | Flop 3 |
|---|---|---|---|
| 90.0% | Dialogue content preserved | ||
| 100.0% | Setting descriptions preserved | ||
| 97.8% | Tense transformation accuracy |