speechless-zephyr-code-functionary-7B: benchmark results
uukuguy's mixture-of-multi-LoRAs merge combining zephyr-7b-beta, functionary-small-v2.2 and speechless-code-mistral-7b-v1.0 LoRA modules on a Mistral 7B base. Provider: uukuguy. Released 2024-01-23. Access: Open.
Aggregate Rating 1384 ± 22, rank #1277 of 1656 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 64.71 | Accuracy (%) (5-shot) | 75.3 |
| Open LLM Leaderboard - MuSR | 42.68 | Score | 67.2 |
| Open LLM Leaderboard v1 - WinoGrande | 78.69 | Accuracy (%) (5-shot) | 64.9 |
| Open LLM Leaderboard v1 - HellaSwag | 83.88 | Normalized accuracy (%) (10-shot) | 62.2 |
| Open Portuguese LLM - OAB Exams | 44.87 | Accuracy (%) | 59.5 |
| Open LLM Leaderboard v1 - GSM8K | 43.82 | Accuracy (%) (5-shot) | 58 |
| Open Portuguese LLM - ENEM | 63.96 | Accuracy (%) | 57.1 |
| Open LLM Leaderboard - GPQA | 30.03 | Score | 56.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 61.52 | Normalized accuracy (%) (25-shot) | 53.2 |
| Open Portuguese LLM - BLUEX | 52.29 | Accuracy (%) | 52.3 |
| Open Portuguese LLM - ASSIN2 RTE | 89.37 | Macro F1 (%) | 49.9 |
| Open Portuguese LLM - FaQuAD NLI | 59.28 | Macro F1 (%) | 43.3 |
Interactive version: theaggregate.ai/model?slug=speechless-zephyr-code-functionary-7b · How It Works · Data refreshed daily, snapshot 2026-10-20.