Voice/dialogue sheets
Extract dialogue from given text as voice sheets.
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 | ||
|---|---|---|---|---|
| Stealth: Healer Alpha | 76% | $0.0000 | 5.0s | |
| DeepSeek V4 Flash | 62% | $0.0001 | 3.7s | |
| Gemini 2.5 Flash Lite | 66% | $0.0001 | 610ms | |
| Mistral Small 3.2 24B | 72% | $0.0001 | 2.1s | |
| Stealth: Hunter Alpha | 70% | $0.0000 | 18.7s | |
| Mistral Small 4 | 60% | $0.0001 | 1.3s | |
| GPT-4o Mini (temp=0) | 60% | $0.0001 | 3.7s | |
| Llama 3.1 8B | 54% | $0.0001 | 919ms | |
| Gemma 4 26B | 88% | $0.0001 | 4.5s | |
| Qwen3 235B A22B Instruct 2507 | 72% | $0.0001 | 4.9s | |
| DeepSeek-V2 Chat | 80% | $0.0001 | 8.2s | |
| Gemma 4 31B | 100% | $0.0001 | 8.3s | |
| Gemma 3 12B | 52% | $0.0000 | 4.1s | |
| DeepSeek V4 Flash (Reasoning) | 78% | $0.0002 | 20.6s | |
| Z.AI GLM 4.5 | 76% | $0.0004 | 5.3s | |
| Hermes 3 405B | 58% | $0.0000 | 13.3s | |
| Gemini 3.1 Flash Lite (Reasoning) | 78% | $0.0003 | 1.4s | |
| Gemini 3.1 Flash Lite (Preview) | 70% | $0.0003 | 979ms | |
| DeepSeek V3 (2025-03-24) | 84% | $0.0003 | 6.1s | |
| ByteDance Seed 1.6 Flash | 66% | $0.0002 | 4.5s | |
Cost vs Performance
Compares total cost for this test against the test score. Quadrant lines are drawn at the median values. Only models with available cost data are shown.
2 low-scoring outliers hidden: Ministral 3 3B (0.0%), LFM2 24B (0.0%).
Most Stable Models (Top 20)
Ranked by stability (median × consistency). Click a model name to view its detail page.
| Score | Consistency | Stability | ||
|---|---|---|---|---|
| Claude Opus 4.6 (Reasoning) | 100% | 100% | 100% | |
| Claude Opus 4.6 | 100% | 100% | 100% | |
| Claude Sonnet 4.6 | 100% | 100% | 100% | |
| Claude Sonnet 4 | 100% | 100% | 100% | |
| Gemma 4 31B | 100% | 100% | 100% | |
| Claude 3.7 Sonnet | 100% | 100% | 100% | |
| Qwen3.6 Max Preview | 96% | 61% | 61% | |
| GPT-4.1 | 96% | 61% | 61% | |
| GPT-4o, May 13th (temp=0) | 96% | 61% | 61% | |
| ByteDance Seed 1.6 | 94% | 53% | 53% | |
| Gemma 4 31B (Reasoning) | 92% | 46% | 46% | |
| Grok 4.20 (Reasoning) | 92% | 46% | 46% | |
| GPT-4o, Aug. 6th (temp=0) | 92% | 46% | 46% | |
| Claude Sonnet 4.6 (Reasoning) | 90% | 40% | 40% | |
| GPT-4o, May 13th (temp=1) | 90% | 40% | 40% | |
| Gemma 4 26B | 88% | 35% | 35% | |
| Qwen 3.5 122B | 88% | 35% | 35% | |
| Gemini 2.5 Flash | 88% | 35% | 35% | |
| Claude Opus 4 | 86% | 31% | 31% | |
| Qwen 3.5 Plus (2026-02-15) | 86% | 31% | 31% | |
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 | ||
|---|---|---|---|---|---|
| Gemma 4 31B | 100% | $0.0001 | 8.3s | 100% | |
| Claude Sonnet 4.6 | 100% | $0.0033 | 1.9s | 100% | |
| Claude Sonnet 4 | 100% | $0.0033 | 2.8s | 100% | |
| Claude 3.7 Sonnet | 100% | $0.0034 | 3.5s | 100% | |
| Claude Opus 4.6 | 100% | $0.0055 | 3.9s | 100% | |
| Claude Opus 4.6 (Reasoning) | 100% | $0.0060 | 3.1s | 100% | |
| GPT-4.1 | 96% | $0.0017 | 2.6s | 61% | |
| GPT-4o, May 13th (temp=0) | 96% | $0.0037 | 5.2s | 61% | |
| GPT-4o, Aug. 6th (temp=0) | 92% | $0.0022 | 2.4s | 46% | |
| Gemini 2.5 Flash | 88% | $0.0005 | 933ms | 35% | |
| ByteDance Seed 1.6 | 94% | $0.0013 | 14.2s | 53% | |
| Gemma 4 26B | 88% | $0.0001 | 4.5s | 35% | |
| Qwen 3.5 Plus (2026-02-15) | 86% | $0.0005 | 6.6s | 31% | |
| Grok 4 Fast | 84% | $0.0003 | 3.6s | 27% | |
| Grok 4.1 Fast | 84% | $0.0004 | 4.4s | 27% | |
| GPT-4o, May 13th (temp=1) | 90% | $0.0037 | 5.2s | 40% | |
| DeepSeek V3 (2025-03-24) | 84% | $0.0003 | 6.1s | 27% | |
| Grok 4.20 (Reasoning) | 92% | $0.0035 | 12.9s | 46% | |
| Claude Haiku 4.5 | 82% | $0.0011 | 1.6s | 23% | |
| Gemini 3 Flash (Preview) | 80% | $0.0006 | 1.8s | 20% | |
| Model | Total â–¼ | Simple | Simple (1-shot) | Simple (5-shot) | Multiple speakers | Unattributed dialogue |
|---|---|---|---|---|---|---|
| Claude Opus 4.6 (Reasoning) | 100% | 100% | 100% | 100% | 100% | 100% |
| Claude Opus 4.6 | 100% | 100% | 100% | 100% | 100% | 100% |
| Claude Sonnet 4.6 | 100% | 100% | 100% | 100% | 100% | 100% |
| Claude Sonnet 4 | 100% | 100% | 100% | 100% | 100% | 100% |
| Gemma 4 31B | 100% | 100% | 100% | 100% | 100% | 100% |
| Claude 3.7 Sonnet | 100% | 100% | 100% | 100% | 100% | 100% |
| Qwen3.6 Max Preview | 96% | 90% | 100% | 100% | 90% | 100% |
| GPT-4.1 | 96% | 90% | 100% | 100% | 90% | 100% |
| GPT-4o, May 13th (temp=0) | 96% | 100% | 100% | 100% | 80% | 100% |
| ByteDance Seed 1.6 | 94% | 90% | 100% | 100% | 80% | 100% |
| Gemma 4 31B (Reasoning) | 92% | 90% | 90% | 100% | 80% | 100% |
| Grok 4.20 (Reasoning) | 92% | 100% | 100% | 100% | 60% | 100% |
| GPT-4o, Aug. 6th (temp=0) | 92% | 100% | 100% | 100% | 60% | 100% |
| Claude Sonnet 4.6 (Reasoning) | 90% | 100% | 90% | 60% | 100% | 100% |
| GPT-4o, May 13th (temp=1) | 90% | 90% | 100% | 70% | 90% | 100% |
1–15 of 147
Page 1 / 10
Simple
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 | 90% | $0.0000 | 446ms | |
| Mistral Small 4 | 90% | $0.0001 | 1.2s | |
| Gemini 3.1 Flash Lite | 90% | $0.0001 | 772ms | |
| Gemini 3.1 Flash Lite (Preview) | 100% | $0.0001 | 808ms | |
| Gemini 3.1 Flash Lite (Reasoning) | 90% | $0.0001 | 871ms | |
| Qwen3 235B A22B Instruct 2507 | 90% | $0.0000 | 1.7s | |
| DeepSeek V4 Flash | 100% | $0.0000 | 6.6s | |
| Gemini 2.5 Flash | 60% | $0.0002 | 697ms | |
| Llama 3.1 70B | 70% | $0.0002 | 1.1s | |
| Gemma 4 26B | 100% | $0.0000 | 3.3s | |
| Mistral Large 3 | 100% | $0.0001 | 1.4s | |
| GPT-4.1 Mini | 90% | $0.0001 | 2.4s | |
| Mistral Small 3.2 24B | 60% | $0.0000 | 1.6s | |
| DeepSeek V3 (2024-12-26) | 100% | $0.0001 | 2.4s | |
| Grok 4.20 | 100% | $0.0003 | 871ms | |
| Grok 4 Fast | 80% | $0.0002 | 2.3s | |
| Grok 4.20 (Beta) | 100% | $0.0005 | 613ms | |
| Gemma 4 31B | 100% | $0.0000 | 4.2s | |
| GPT-4o Mini (temp=0) | 100% | $0.0001 | 3.1s | |
| Z.AI GLM 4.5 | 100% | $0.0001 | 2.7s | |
Most Stable Models (Top 20)
Ranked by stability (median × consistency). Click a model name to view its detail page.
| Score | Consistency | Stability | ||
|---|---|---|---|---|
| Claude Opus 4.6 (Reasoning) | 100% | 100% | 100% | |
| Claude Sonnet 4.6 (Reasoning) | 100% | 100% | 100% | |
| GPT-5.5 (Reasoning, Low) | 100% | 100% | 100% | |
| Claude Opus 4.6 | 100% | 100% | 100% | |
| Grok 4.20 (Reasoning) | 100% | 100% | 100% | |
| Claude Sonnet 4.6 | 100% | 100% | 100% | |
| Claude Sonnet 4 | 100% | 100% | 100% | |
| Claude Opus 4 | 100% | 100% | 100% | |
| Gemma 4 31B | 100% | 100% | 100% | |
| Z.AI GLM 4.5 | 100% | 100% | 100% | |
| Qwen 3.5 Plus (2026-02-15) | 100% | 100% | 100% | |
| Gemini 3.1 Flash Lite (Preview) | 100% | 100% | 100% | |
| Gemma 4 26B | 100% | 100% | 100% | |
| Mistral Large 3 | 100% | 100% | 100% | |
| GPT-4o, May 13th (temp=0) | 100% | 100% | 100% | |
| DeepSeek-V2 Chat | 100% | 100% | 100% | |
| Claude 3.5 Sonnet | 100% | 100% | 100% | |
| Grok 4.20 (Beta) | 100% | 100% | 100% | |
| DeepSeek V3 (2024-12-26) | 100% | 100% | 100% | |
| Claude 3.7 Sonnet | 100% | 100% | 100% | |
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 | ||
|---|---|---|---|---|---|
| Gemini 3.1 Flash Lite (Preview) | 100% | $0.0001 | 808ms | 100% | |
| Mistral Large 3 | 100% | $0.0001 | 1.4s | 100% | |
| Grok 4.20 | 100% | $0.0003 | 871ms | 100% | |
| Grok 4.20 (Beta) | 100% | $0.0005 | 613ms | 100% | |
| DeepSeek V3 (2024-12-26) | 100% | $0.0001 | 2.4s | 100% | |
| Z.AI GLM 4.5 | 100% | $0.0001 | 2.7s | 100% | |
| GPT-4o Mini (temp=0) | 100% | $0.0001 | 3.1s | 100% | |
| Gemma 4 26B | 100% | $0.0000 | 3.3s | 100% | |
| Qwen 3.5 Plus (2026-02-15) | 100% | $0.0002 | 3.6s | 100% | |
| Gemma 4 31B | 100% | $0.0000 | 4.2s | 100% | |
| DeepSeek-V2 Chat | 100% | $0.0000 | 4.3s | 100% | |
| GPT-4o, Aug. 6th (temp=0) | 100% | $0.0009 | 1.5s | 100% | |
| GPT-4o, Aug. 6th (temp=1) | 100% | $0.0009 | 1.5s | 100% | |
| Claude Sonnet 4.6 | 100% | $0.0013 | 1.4s | 100% | |
| Claude Sonnet 4 | 100% | $0.0013 | 2.1s | 100% | |
| DeepSeek V4 Flash | 100% | $0.0000 | 6.6s | 100% | |
| Writer: Palmyra X5 | 100% | $0.0004 | 5.9s | 100% | |
| Claude 3.7 Sonnet | 100% | $0.0016 | 2.3s | 100% | |
| GPT-4o, May 13th (temp=0) | 100% | $0.0014 | 3.9s | 100% | |
| Claude 3.5 Sonnet | 100% | $0.0015 | 4.0s | 100% | |
| Median | Evaluator | Top 3 | Flop 3 |
|---|---|---|---|
| 60.0% | Matches Regex |
Simple (1-shot)
1-shot Rule Following
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 | ||
|---|---|---|---|---|
| Gemma 3 12B | 100% | $0.0000 | 1.9s | |
| Mistral Small Creative | 90% | $0.0000 | 877ms | |
| Ministral 3 8B | 100% | $0.0000 | 644ms | |
| Stealth: Healer Alpha | 100% | $0.0000 | 3.5s | |
| Gemini 2.5 Flash Lite | 90% | $0.0000 | 451ms | |
| DeepSeek V4 Flash | 100% | $0.0000 | 1.4s | |
| Mistral Small 3.2 24B | 100% | $0.0000 | 1.2s | |
| GPT-4.1 Nano | 100% | $0.0000 | 1.6s | |
| Mistral Small 4 | 100% | $0.0001 | 1.1s | |
| Llama 3.1 8B | 60% | $0.0001 | 801ms | |
| Gemma 4 31B | 100% | $0.0001 | 4.7s | |
| GPT-4o Mini (temp=0) | 100% | $0.0001 | 3.2s | |
| DeepSeek-V2 Chat | 100% | $0.0000 | 4.4s | |
| Qwen3 235B A22B Instruct 2507 | 100% | $0.0001 | 4.5s | |
| DeepSeek V4 Flash (Reasoning) | 100% | $0.0001 | 5.5s | |
| GPT-5.4 Nano (Reasoning, Low) | 70% | $0.0001 | 1.0s | |
| Gemini 3.1 Flash Lite (Reasoning) | 90% | $0.0001 | 778ms | |
| Gemini 3.1 Flash Lite | 80% | $0.0001 | 920ms | |
| Gemini 3.1 Flash Lite (Preview) | 70% | $0.0001 | 801ms | |
| Claude 3 Haiku | 100% | $0.0001 | 3.6s | |
Most Stable Models (Top 20)
Ranked by stability (median × consistency). Click a model name to view its detail page.
| Score | Consistency | Stability | ||
|---|---|---|---|---|
| Claude Opus 4.6 (Reasoning) | 100% | 100% | 100% | |
| Qwen3.6 Max Preview | 100% | 100% | 100% | |
| Z.AI GLM 5 Turbo | 100% | 100% | 100% | |
| Grok 4.3 (Reasoning) | 100% | 100% | 100% | |
| GPT-5.4 (Reasoning) | 100% | 100% | 100% | |
| Claude Opus 4.7 (Reasoning) | 100% | 100% | 100% | |
| GPT-5.5 (Reasoning) | 100% | 100% | 100% | |
| GPT-5 Mini | 100% | 100% | 100% | |
| GPT-5.5 (Reasoning, Low) | 100% | 100% | 100% | |
| GPT-5.1 | 100% | 100% | 100% | |
| Claude Opus 4.6 | 100% | 100% | 100% | |
| GPT-5 | 100% | 100% | 100% | |
| Qwen 3.5 397B A17B | 100% | 100% | 100% | |
| Grok 4.20 (Beta, Reasoning) | 100% | 100% | 100% | |
| Grok 4.20 (Reasoning) | 100% | 100% | 100% | |
| Claude Sonnet 4.6 | 100% | 100% | 100% | |
| ByteDance Seed 1.6 | 100% | 100% | 100% | |
| o4 Mini High | 100% | 100% | 100% | |
| DeepSeek V4 Pro (Reasoning) | 100% | 100% | 100% | |
| Claude Opus 4.7 | 100% | 100% | 100% | |
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 | ||
|---|---|---|---|---|---|
| Ministral 3 8B | 100% | $0.0000 | 644ms | 100% | |
| Mistral Small 3.2 24B | 100% | $0.0000 | 1.2s | 100% | |
| DeepSeek V4 Flash | 100% | $0.0000 | 1.4s | 100% | |
| Gemma 3 12B | 100% | $0.0000 | 1.9s | 100% | |
| Mistral Small 4 | 100% | $0.0001 | 1.1s | 100% | |
| GPT-4.1 Nano | 100% | $0.0000 | 1.6s | 100% | |
| Stealth: Healer Alpha | 100% | $0.0000 | 3.5s | 100% | |
| GPT-4o Mini (temp=0) | 100% | $0.0001 | 3.2s | 100% | |
| DeepSeek-V2 Chat | 100% | $0.0000 | 4.4s | 100% | |
| Gemini 2.5 Flash | 100% | $0.0002 | 666ms | 100% | |
| DeepSeek V3 (2025-03-24) | 100% | $0.0001 | 3.0s | 100% | |
| Qwen3 235B A22B Instruct 2507 | 100% | $0.0001 | 4.5s | 100% | |
| Gemma 4 31B | 100% | $0.0001 | 4.7s | 100% | |
| Hermes 3 70B | 100% | $0.0001 | 3.5s | 100% | |
| GPT-4.1 Mini | 100% | $0.0002 | 2.3s | 100% | |
| Claude 3 Haiku | 100% | $0.0001 | 3.6s | 100% | |
| Z.AI GLM 4.5 | 100% | $0.0002 | 2.7s | 100% | |
| DeepSeek V4 Flash (Reasoning) | 100% | $0.0001 | 5.5s | 100% | |
| WizardLM 2 8x22b | 100% | $0.0002 | 4.5s | 100% | |
| Grok 4.20 | 100% | $0.0004 | 1.0s | 100% | |
| Median | Evaluator | Top 3 | Flop 3 |
|---|---|---|---|
| 90.0% | Matches Regex |
Simple (5-shot)
Few-shot Rule Following
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 | ||
|---|---|---|---|---|
| Gemma 3 4B | 100% | $0.0000 | 929ms | |
| Mistral Small Creative | 100% | $0.0001 | 653ms | |
| Gemini 2.5 Flash Lite | 70% | $0.0001 | 493ms | |
| Ministral 3 8B | 100% | $0.0001 | 633ms | |
| Mistral Small 3.2 24B | 100% | $0.0001 | 1.2s | |
| Mistral Small 4 | 100% | $0.0001 | 844ms | |
| Llama 3.1 8B | 80% | $0.0002 | 739ms | |
| Ministral 3 14B | 100% | $0.0002 | 779ms | |
| DeepSeek V4 Flash | 70% | $0.0000 | 1.5s | |
| GPT-4o Mini (temp=0) | 100% | $0.0002 | 959ms | |
| Gemma 3 12B | 100% | $0.0000 | 1.7s | |
| GPT-4o Mini (temp=1) | 100% | $0.0002 | 977ms | |
| Mistral NeMO | 60% | $0.0001 | 1.3s | |
| Gemma 4 26B | 90% | $0.0001 | 4.3s | |
| GPT-4.1 Nano | 100% | $0.0001 | 1.8s | |
| GPT-5.4 Nano | 100% | $0.0002 | 1.1s | |
| GPT-5.4 Nano (Reasoning, Low) | 100% | $0.0002 | 1.2s | |
| Gemini 3.1 Flash Lite (Preview) | 90% | $0.0003 | 754ms | |
| Gemini 3.1 Flash Lite (Reasoning) | 100% | $0.0003 | 838ms | |
| Gemini 3.1 Flash Lite | 80% | $0.0003 | 873ms | |
Most Stable Models (Top 20)
Ranked by stability (median × consistency). Click a model name to view its detail page.
| Score | Consistency | Stability | ||
|---|---|---|---|---|
| Claude Opus 4.6 (Reasoning) | 100% | 100% | 100% | |
| Qwen3.6 Max Preview | 100% | 100% | 100% | |
| Z.AI GLM 5 Turbo | 100% | 100% | 100% | |
| Grok 4.3 (Reasoning) | 100% | 100% | 100% | |
| GPT-5.4 (Reasoning) | 100% | 100% | 100% | |
| Claude Opus 4.7 (Reasoning) | 100% | 100% | 100% | |
| GPT-5.5 (Reasoning) | 100% | 100% | 100% | |
| GPT-5.1 | 100% | 100% | 100% | |
| Claude Opus 4.6 | 100% | 100% | 100% | |
| Qwen 3.5 397B A17B | 100% | 100% | 100% | |
| Gemma 4 31B (Reasoning) | 100% | 100% | 100% | |
| Qwen 3.5 122B | 100% | 100% | 100% | |
| Grok 4.20 (Beta, Reasoning) | 100% | 100% | 100% | |
| GPT-5.4 (Reasoning, Low) | 100% | 100% | 100% | |
| Grok 4.20 (Reasoning) | 100% | 100% | 100% | |
| Claude Sonnet 4.6 | 100% | 100% | 100% | |
| ByteDance Seed 1.6 | 100% | 100% | 100% | |
| GPT-5.4 Mini (Reasoning) | 100% | 100% | 100% | |
| Claude Opus 4.7 | 100% | 100% | 100% | |
| Claude Opus 4.5 | 100% | 100% | 100% | |
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 Creative | 100% | $0.0001 | 653ms | 100% | |
| Gemma 3 4B | 100% | $0.0000 | 929ms | 100% | |
| Ministral 3 8B | 100% | $0.0001 | 633ms | 100% | |
| Ministral 3 14B | 100% | $0.0002 | 779ms | 100% | |
| Mistral Small 4 | 100% | $0.0001 | 844ms | 100% | |
| GPT-4o Mini (temp=0) | 100% | $0.0002 | 959ms | 100% | |
| GPT-4o Mini (temp=1) | 100% | $0.0002 | 977ms | 100% | |
| Mistral Small 3.2 24B | 100% | $0.0001 | 1.2s | 100% | |
| Gemini 3.1 Flash Lite (Reasoning) | 100% | $0.0003 | 838ms | 100% | |
| Gemma 3 12B | 100% | $0.0000 | 1.7s | 100% | |
| GPT-5.4 Nano | 100% | $0.0002 | 1.1s | 100% | |
| GPT-5.4 Nano (Reasoning, Low) | 100% | $0.0002 | 1.2s | 100% | |
| Claude 3 Haiku | 100% | $0.0003 | 1.1s | 100% | |
| GPT-4.1 Nano | 100% | $0.0001 | 1.8s | 100% | |
| Mistral Large 3 | 100% | $0.0005 | 1.4s | 100% | |
| Arcee AI: Trinity Large (Preview) | 100% | $0.0000 | 3.1s | 100% | |
| Grok 4.20 (Beta) | 100% | $0.0007 | 601ms | 100% | |
| GPT-4.1 Mini | 100% | $0.0004 | 1.9s | 100% | |
| Llama 3.1 70B | 100% | $0.0008 | 993ms | 100% | |
| Z.AI GLM 4.5 | 100% | $0.0003 | 2.8s | 100% | |
| Median | Evaluator | Top 3 | Flop 3 |
|---|---|---|---|
| 90.0% | Matches Regex |
Multiple speakers
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 | ||
|---|---|---|---|---|
| Stealth: Hunter Alpha | 60% | $0.0000 | 15.5s | |
| Stealth: Healer Alpha | 60% | $0.0000 | 4.7s | |
| Gemma 4 26B | 100% | $0.0002 | 6.9s | |
| Gemma 4 31B | 100% | $0.0002 | 17.6s | |
| DeepSeek V3.2 | 80% | $0.0002 | 9.0s | |
| Gemini 2.5 Flash Lite (Reasoning) | 90% | $0.0005 | 3.8s | |
| GPT-4.1 Mini | 90% | $0.0006 | 3.8s | |
| Grok 4.1 Fast | 60% | $0.0005 | 5.3s | |
| Gemini 2.5 Flash | 100% | $0.0008 | 1.3s | |
| Grok 4 Fast | 60% | $0.0005 | 5.5s | |
| DeepSeek V3.1 | 70% | $0.0003 | 13.7s | |
| Gemini 3 Flash (Preview) | 100% | $0.0010 | 2.0s | |
| Gemini 2.5 Flash (Reasoning) | 60% | $0.0017 | 3.1s | |
| Qwen 3.5 Plus (2026-02-15) | 100% | $0.0008 | 9.9s | |
| Gemma 4 26B (Reasoning) | 90% | $0.0004 | 33.5s | |
| Xiaomi MIMO v2.5 Pro | 90% | $0.0021 | 10.5s | |
| Claude Haiku 4.5 | 100% | $0.0018 | 2.1s | |
| GPT-4.1 | 90% | $0.0028 | 3.3s | |
| ByteDance Seed 1.6 | 80% | $0.0024 | 25.7s | |
| Mistral Large 2 | 100% | $0.0029 | 8.3s | |
Most Stable Models (Top 20)
Ranked by stability (median × consistency). Click a model name to view its detail page.
| Score | Consistency | Stability | ||
|---|---|---|---|---|
| Claude Opus 4.6 (Reasoning) | 100% | 100% | 100% | |
| Gemini 3.1 Pro (Preview) | 100% | 100% | 100% | |
| Claude Sonnet 4.6 (Reasoning) | 100% | 100% | 100% | |
| Claude Opus 4.7 (Reasoning) | 100% | 100% | 100% | |
| Claude Opus 4.6 | 100% | 100% | 100% | |
| Claude Sonnet 4.6 | 100% | 100% | 100% | |
| Claude Opus 4.7 | 100% | 100% | 100% | |
| Gemini 3 Pro (Preview) | 100% | 100% | 100% | |
| Claude Sonnet 4 | 100% | 100% | 100% | |
| Claude Sonnet 4.5 | 100% | 100% | 100% | |
| Claude Opus 4 | 100% | 100% | 100% | |
| Gemma 4 31B | 100% | 100% | 100% | |
| Qwen 3.5 Plus (2026-02-15) | 100% | 100% | 100% | |
| Gemma 4 26B | 100% | 100% | 100% | |
| Gemini 3 Flash (Preview) | 100% | 100% | 100% | |
| Claude Haiku 4.5 | 100% | 100% | 100% | |
| ByteDance Seed 2.0 Lite | 100% | 100% | 100% | |
| Claude 3.7 Sonnet | 100% | 100% | 100% | |
| Mistral Large 2 | 100% | 100% | 100% | |
| Gemini 2.5 Flash | 100% | 100% | 100% | |
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 | ||
|---|---|---|---|---|---|
| Gemini 2.5 Flash | 100% | $0.0008 | 1.3s | 100% | |
| Gemma 4 26B | 100% | $0.0002 | 6.9s | 100% | |
| Gemini 3 Flash (Preview) | 100% | $0.0010 | 2.0s | 100% | |
| Claude Haiku 4.5 | 100% | $0.0018 | 2.1s | 100% | |
| Qwen 3.5 Plus (2026-02-15) | 100% | $0.0008 | 9.9s | 100% | |
| Gemma 4 31B | 100% | $0.0002 | 17.6s | 100% | |
| Mistral Large 2 | 100% | $0.0029 | 8.3s | 100% | |
| Claude Sonnet 4.6 | 100% | $0.0054 | 2.6s | 100% | |
| Claude Sonnet 4.5 | 100% | $0.0054 | 3.5s | 100% | |
| Claude Sonnet 4 | 100% | $0.0054 | 4.0s | 100% | |
| Claude 3.7 Sonnet | 100% | $0.0055 | 4.7s | 100% | |
| Claude Sonnet 4.6 (Reasoning) | 100% | $0.0060 | 3.2s | 100% | |
| ByteDance Seed 2.0 Lite | 100% | $0.0029 | 33.8s | 100% | |
| Claude Opus 4.6 | 100% | $0.0091 | 5.0s | 100% | |
| Claude Opus 4.6 (Reasoning) | 100% | $0.0098 | 4.1s | 100% | |
| Claude Opus 4.7 | 100% | $0.013 | 2.7s | 100% | |
| Claude Opus 4.7 (Reasoning) | 100% | $0.013 | 2.7s | 100% | |
| Gemini 2.5 Flash Lite (Reasoning) | 90% | $0.0005 | 3.8s | 40% | |
| GPT-4.1 Mini | 90% | $0.0006 | 3.8s | 40% | |
| GPT-4.1 | 90% | $0.0028 | 3.3s | 40% | |
| Median | Evaluator | Top 3 | Flop 3 |
|---|---|---|---|
| 20.0% | Matches Regex |
Unattributed dialogue
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 | ||
|---|---|---|---|---|
| Stealth: Aurora Alpha | 100% | — | 971ms | |
| Llama 3.1 8B | 90% | $0.0001 | 1.2s | |
| Gemini 2.5 Flash Lite | 70% | $0.0001 | 781ms | |
| Inception Mercury | 100% | $0.0002 | 1.0s | |
| Mistral Small 3.2 24B | 100% | $0.0001 | 3.5s | |
| Stealth: Healer Alpha | 90% | $0.0000 | 5.6s | |
| Gemma 4 26B | 100% | $0.0001 | 4.5s | |
| Inception Mercury 2 | 100% | $0.0005 | 815ms | |
| Grok 4 Fast | 100% | $0.0003 | 3.2s | |
| Nemotron 3 Super | 100% | $0.0000 | 10.6s | |
| Gemini 3.1 Flash Lite (Reasoning) | 100% | $0.0005 | 2.4s | |
| Gemini 3.1 Flash Lite (Preview) | 60% | $0.0005 | 1.2s | |
| Gemini 3.1 Flash Lite | 100% | $0.0005 | 1.8s | |
| ByteDance Seed 1.6 Flash | 70% | $0.0003 | 5.0s | |
| Gemma 3 27B | 100% | $0.0001 | 8.6s | |
| Grok 4.1 Fast | 100% | $0.0003 | 4.0s | |
| DeepSeek V4 Flash (Reasoning) | 100% | $0.0002 | 9.1s | |
| Gemma 4 31B | 100% | $0.0001 | 11.6s | |
| Llama 3.1 70B | 100% | $0.0005 | 2.3s | |
| Gemini 2.5 Flash Lite (Reasoning) | 100% | $0.0004 | 3.4s | |
Most Stable Models (Top 20)
Ranked by stability (median × consistency). Click a model name to view its detail page.
| Score | Consistency | Stability | ||
|---|---|---|---|---|
| Claude Opus 4.6 (Reasoning) | 100% | 100% | 100% | |
| Qwen3.6 Max Preview | 100% | 100% | 100% | |
| Gemini 3.1 Pro (Preview) | 100% | 100% | 100% | |
| Z.AI GLM 5 Turbo | 100% | 100% | 100% | |
| Claude Sonnet 4.6 (Reasoning) | 100% | 100% | 100% | |
| Grok 4.3 (Reasoning) | 100% | 100% | 100% | |
| GPT-5.4 (Reasoning) | 100% | 100% | 100% | |
| Claude Opus 4.7 (Reasoning) | 100% | 100% | 100% | |
| GPT-5.5 (Reasoning) | 100% | 100% | 100% | |
| GPT-5 Mini | 100% | 100% | 100% | |
| GPT-5.5 (Reasoning, Low) | 100% | 100% | 100% | |
| GPT-5.1 | 100% | 100% | 100% | |
| Claude Opus 4.6 | 100% | 100% | 100% | |
| MoonshotAI: Kimi K2.6 | 100% | 100% | 100% | |
| GPT-5 | 100% | 100% | 100% | |
| Qwen 3.5 397B A17B | 100% | 100% | 100% | |
| Gemma 4 31B (Reasoning) | 100% | 100% | 100% | |
| Qwen 3.5 122B | 100% | 100% | 100% | |
| Qwen 3.5 Plus (2026-04-20) | 100% | 100% | 100% | |
| Gemma 4 26B (Reasoning) | 100% | 100% | 100% | |
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 | ||
|---|---|---|---|---|---|
| Stealth: Aurora Alpha | 100% | — | 971ms | 100% | |
| Inception Mercury | 100% | $0.0002 | 1.0s | 100% | |
| Inception Mercury 2 | 100% | $0.0005 | 815ms | 100% | |
| Gemini 3.1 Flash Lite | 100% | $0.0005 | 1.8s | 100% | |
| Llama 3.1 70B | 100% | $0.0005 | 2.3s | 100% | |
| Gemini 3.1 Flash Lite (Reasoning) | 100% | $0.0005 | 2.4s | 100% | |
| Mistral Small 3.2 24B | 100% | $0.0001 | 3.5s | 100% | |
| Grok 4 Fast | 100% | $0.0003 | 3.2s | 100% | |
| Gemini 2.5 Flash Lite (Reasoning) | 100% | $0.0004 | 3.4s | 100% | |
| Gemma 4 26B | 100% | $0.0001 | 4.5s | 100% | |
| Grok 4.1 Fast | 100% | $0.0003 | 4.0s | 100% | |
| Gemini 3 Flash (Preview) | 100% | $0.0009 | 2.6s | 100% | |
| Mistral Large 3 | 100% | $0.0005 | 4.0s | 100% | |
| Claude Haiku 4.5 | 100% | $0.0016 | 1.9s | 100% | |
| GPT-5.4 Mini (Reasoning) | 100% | $0.0020 | 3.0s | 100% | |
| Gemma 3 27B | 100% | $0.0001 | 8.6s | 100% | |
| Qwen 2.5 72B | 100% | $0.0002 | 8.2s | 100% | |
| DeepSeek V4 Flash (Reasoning) | 100% | $0.0002 | 9.1s | 100% | |
| Gemini 2.5 Flash (Reasoning) | 100% | $0.0023 | 3.6s | 100% | |
| MiniMax M2.5 | 100% | $0.0008 | 8.3s | 100% | |
| Median | Evaluator | Top 3 | Flop 3 |
|---|---|---|---|
| 100.0% | Matches Regex |