NeuralExperiment-7B-MagicCoder-v7.5: benchmark results
Kukedlc's math-, code-, and logic-focused fine-tune of his Mistral-based NeuralExperiment-7b-dare-ties merge, trained on Orca-Math and Magicoder data. Provider: Kukedlc. Released 2024-03-07. Access: Open.
Aggregate Rating 1454 ± 22, rank #975 of 1656 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 72.11 | MC2 (%) (0-shot) | 93.1 |
| Open LLM Leaderboard v1 - WinoGrande | 83.5 | Accuracy (%) (5-shot) | 89.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 71.33 | Normalized accuracy (%) (25-shot) | 89 |
| Open Chinese LLM - TruthfulQA MC | 61.64 | Accuracy (%) | 88.8 |
| Open LLM Leaderboard v1 - HellaSwag | 87.94 | Normalized accuracy (%) (10-shot) | 88.8 |
| Open LLM Leaderboard v1 - GSM8K | 66.19 | Accuracy (%) (5-shot) | 81.3 |
| Open Portuguese LLM - FaQuAD NLI | 76.41 | Macro F1 (%) | 76.3 |
| Open Portuguese LLM - ASSIN2 RTE | 92.07 | Macro F1 (%) | 73.9 |
| Open LLM Leaderboard v1 - MMLU | 64.62 | Accuracy (%) (5-shot) | 73.5 |
| Open LLM Leaderboard - MuSR | 42.82 | Score | 68.8 |
| Open Chinese LLM - GSM8K | 47.61 | Accuracy (%) | 59.2 |
| Open Portuguese LLM - ENEM | 64.31 | Accuracy (%) | 58.2 |
Interactive version: theaggregate.ai/model?slug=neuralexperiment-7b-magiccoder-v7-5 · How It Works · Data refreshed daily, snapshot 2026-10-20.