TouchstoneGPT-7B-Instruct: benchmark results
Provider: Other.
Aggregate Rating 1331 ± 52, rank #8070 of 10552 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| FinReasoning (Deep Insight) - Justification and Causal Depth | 34.2 | Rubric score (0-100): a judge score fused from DeepSeek-V3 a | 5.6 |
| Fin-RATE | 0.41 | Macro Accuracy (imported) | 0 |
| FinReasoning (Data Alignment) | 18.1 | Success-weighted score (0-100): the success rate (items whos | 0 |
| FinReasoning (Data Alignment) - Calculation | 17.9 | Success-weighted score (0-100): the success rate (items whos | 0 |
| FinReasoning (Data Alignment) - Reasoning | 12.6 | Success-weighted score (0-100): the success rate (items whos | 0 |
| FinReasoning (Data Alignment) - Verification | 23.8 | Success-weighted score (0-100): the success rate (items whos | 0 |
| FinReasoning (Deep Insight) | 33 | Rubric score (0-100): a judge score fused from DeepSeek-V3 a | 0 |
| FinReasoning (Deep Insight) - Architectural Richness and Strictness | 38.3 | Rubric score (0-100): a judge score fused from DeepSeek-V3 a | 0 |
| FinReasoning (Deep Insight) - Completeness and Comparative Analysis | 30.8 | Rubric score (0-100): a judge score fused from DeepSeek-V3 a | 0 |
| FinReasoning (Deep Insight) - Factuality and Contextualization | 28.7 | Rubric score (0-100): a judge score fused from DeepSeek-V3 a | 0 |
| FinReasoning (Semantic Consistency) | 11.8 | Composite score (0-100): the mean of sentence-level F1 for l | 0 |
| FinReasoning (Semantic Consistency) - Fact | 12 | Composite score (0-100): the mean of sentence-level F1 for l | 0 |
Interactive version: theaggregate.ai/model?slug=touchstonegpt-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-16.