TL;DR
Claude 3.7 Sonnet (anthropic/claude-3.7-sonnet) appears in 4 published studies. 2025-02-24, Claude 3.7 Sonnet and o3-mini building a vague leveraged-ETF strategy: 36.02% for percent change. 2025-07-11, Models generating financial SQL queries, July 2025: 0.667 for median score. 2025-04-10, Seven models generating financial SQL queries: 0.660 for average score. 2025-04-15, Models generating SQL for financial questions, April 2025: 0.660 for average score. Each result belongs to its own task, date and grading method.
- Exact tested ID
- anthropic/claude-3.7-sonnet
- Published studies
- 4
- Provider
- anthropic
Inspect this exact model
Claude 3.7 Sonnet and o3-mini building a vague leveraged-ETF strategy
Claude 3.7 Sonnet: source evidence dated 2025-07-11.
- Percent change
- 36.02%
- Sharpe ratio
- 1.03
- Sortino ratio
- 1.02
- Max drawdown
- 17.55%
- Average drawdown
- 1.83%
- Num trades
- 31.00
The displayed values belong to the identified model versions and this published task, with their original metric units.
How did this version perform on the job I need?
The first measured row below belongs to this model, not its provider's other models. Open that study to see the other arms and tested input.
2025-02-24: Claude 3.7 Sonnet and o3-mini building a vague leveraged-ETF strategySelect a model for the specific tested task. A profile with one study does not establish performance on other jobs.
Test a strategy with Aurora
Select a model for the specific tested task. A profile with one study does not establish performance on other jobs.
Test a strategy with AuroraKeep reporting dates and units beside the answer, and inspect its original citations.
Claude 3.7 Sonnet and o3-mini building a vague leveraged-ETF strategy
See the complete claude 3.7 sonnet and o3-mini building a vague leveraged-etf strategy
Claude 3.7 Sonnet recorded 36.02% for percent change in the 2025-02-24 publication. 1 ambiguous strategy request, 2 models, one generated strategy each, backtested once.
| Published metric | Value |
|---|---|
| Percent change | 36.02% |
| Sharpe ratio | 1.03 |
| Sortino ratio | 1.02 |
| Max drawdown | 17.55% |
| Average drawdown | 1.83% |
| Num trades | 31.00 |
No matching rows. Clear the filter to see all records.
Claude 3.7 Sonnet and o3-mini building a vague leveraged-ETF strategy: read these measurements
View task, sample and measurement units
1 ambiguous strategy request, 2 models, one generated strategy each, backtested once
Turn an ambiguous plain-English request into a complete nested trading strategy JSON object, then backtest it.
Scoring
No score is assigned. The article compares percent change, Sharpe and Sortino ratios and drawdowns from each backtest, alongside the author's reading of the generated rules. Over the same window, holding SPY returned 20.58% with a 1.06 Sharpe ratio, 1.18 Sortino ratio and 9.42% max drawdown.
Cost units
The article publishes no per-run cost. It lists Claude 3.7 Sonnet at $3 input and $15 output per million tokens and o3-mini at $1.10 input and $4.40 output per million tokens.
Time units
The article publishes no per-model timing. It says o3-mini produced its strategy in less than a minute.
Models generating financial SQL queries, July 2025
Open every published row
Claude 3.7 Sonnet recorded 0.667 for median score in the 2025-07-11 publication. About 88 financial questions per the article text, 11 models.
| Published metric | Value |
|---|---|
| Median score | 0.667 |
| Avg score | 0.565 |
| Success rate | 68.54% |
| Avg time | 45,526 ms |
| Input cost | $3.00/M |
| Output cost | $15.00/M |
No matching rows. Clear the filter to see all records.
Models generating financial SQL queries, July 2025: read these measurements
View task, sample and measurement units
About 88 financial questions per the article text, 11 models
Generate SQL queries that answer financial questions against a financial database.
Scoring
Three LLM judges assign a 0 to 1 quality score that is separate from whether the query executes. The article states the score measures query and answer assessment, not investment returns.
Cost units
List prices per million tokens, from $0.10/M input and $0.40/M output (Gemini 2.0 Flash) to $15.00/M input and $75.00/M output (Claude Opus 4). Total run spend was reported as over $200, with $171.88 on Requesty.
Time units
Average time in milliseconds, from 31,140 ms (Gemini 2.0 Flash) to 74,812 ms (Grok-4).
Seven models generating financial SQL queries
Open every published row
Claude 3.7 Sonnet recorded 0.660 for average score in the 2025-04-10 publication. 7 models on a shared SQL question set, the article does not state the number of questions.
| Published metric | Value |
|---|---|
| Average score | 0.660 |
| Median score | 1.000 |
| Std deviation | 0.455 |
| Success rate | 75.0% |
| Avg execution time (ms) | 2,003 |
| Perfect score (1.0) share | 57.5% |
No matching rows. Clear the filter to see all records.
Seven models generating financial SQL queries: read these measurements
View task, sample and measurement units
7 models on a shared SQL question set, the article does not state the number of questions
Generate SQL queries that answer natural-language financial questions over NexusTrade's stock data.
Scoring
Each query receives a 0 to 1 score. The table reports the distribution in buckets 0.0-0.2, 0.3-0.5, 0.6-0.7, 0.8-0.9 and 1.0. The grader is not described in this article.
Cost units
The article publishes no per-model cost.
Time units
Average execution time in milliseconds per model, from 1,296 ms (Gemini 2.0 Flash) to 26,892 ms (DeepSeek V3). The article does not define whether this includes model generation time.
Models generating SQL for financial questions, April 2025
Open every published row
Claude 3.7 Sonnet recorded 0.660 for average score in the 2025-04-15 publication. 60 financial questions per the article text, 11 models.
| Published metric | Value |
|---|---|
| Average score | 0.660 |
| Median score | 1.000 |
| Std dev | 0.455 |
| Success rate | 75.0% |
| Avg execution | 2,003 ms |
| Input cost | $3.00/M |
| Output cost | $15.00/M |
No matching rows. Clear the filter to see all records.
Models generating SQL for financial questions, April 2025: read these measurements
View task, sample and measurement units
60 financial questions per the article text, 11 models
Generate SQL queries that correctly answer financial analysis questions.
Scoring
The article reports scores and success rates but refers readers to an earlier article for the methodology. The grader is not described in this article.
Cost units
List prices per million tokens, from $0.10/M input for GPT-4.1 Nano and Gemini 2.0 Flash to $3.00/M input and $15.00/M output for Claude 3.7 Sonnet and Grok-3-Beta. No measured run spend is published.
Time units
Average execution time in milliseconds, from 1,296 ms (Gemini 2.0 Flash) to 38,614 ms (GPT-4.1 Nano).
No comments yet.