Qwen3.8-27B-FP8: benchmark results
Provider: Alibaba. Access: Open.
Aggregate Rating 1623 ± 50, rank #806 of 10916 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Coding Agent Leaderboard - SWE-Bench Pro -- Ansible (Claude Code) | 82.3 | Score (%) | 88.9 |
| Coding Agent Leaderboard - SWE-Bench Pro -- Ansible (OpenCode) | 87.5 | Score (%) | 87.5 |
| Coding Agent Leaderboard - SWE-Bench Verified (OpenCode) | 93.6 | Score (%) | 87.5 |
| BenCzechMark - Czech Math Reasoning | 66.03 | Duel Win Score (%) | 79.5 |
| Coding Agent Leaderboard - SWE-Bench Verified (Claude Code) | 82.2 | Score (%) | 77.8 |
| BenCzechMark - Czech Language Understanding | 65.17 | Duel Win Score (%) | 76.9 |
| BenCzechMark - NER | 73.72 | Duel Win Score (%) | 75.6 |
| BenCzechMark - Reading Comprehension | 64.96 | Duel Win Score (%) | 74.4 |
| BenCzechMark | 62.62 | Average Duel Win Score (%) | 69.2 |
| BenCzechMark - Factual Knowledge | 59.71 | Duel Win Score (%) | 69.2 |
| BenCzechMark - NLI | 58.09 | Duel Win Score (%) | 69.2 |
| BenCzechMark - Sentiment | 50.64 | Duel Win Score (%) | 42.3 |
Interactive version: theaggregate.ai/model?slug=qwen3-8-27b-fp8 · How It Works · Data refreshed daily, snapshot 2026-11-08.