Qwen2.5 7B (local): real-world results
Qwen2.5 7B (local) passed 1 of 15 runs (7%) on 15 different use cases.
Average cost per run was $0 (local), average wall time 33s, with 0.67 human step-ins per run.
Results by category
| Category | Runs | Passed |
|---|---|---|
| Software & DevOps | 4 | 1/4 |
| Analytics / Metrics | 1 | 0/1 |
| DevOps / Configuration | 1 | 0/1 |
| Finance / Accounts Payable | 1 | 0/1 |
| Finance / Expense Audit | 1 | 0/1 |
| Operations / Inventory | 1 | 0/1 |
| Personal Assistant / Voice Handoff | 1 | 0/1 |
| Privacy / Data Operations | 1 | 0/1 |
| Robotics / Spatial Planning | 1 | 0/1 |
| Scheduling / Calendar | 1 | 0/1 |
| Security, Science & Hardware | 1 | 0/1 |
| Software / Refactor | 1 | 0/1 |
Head-to-head comparisons
- Qwen2.5 7B (local) vs Claude Haiku 5.5
- Qwen2.5 7B (local) vs Claude Opus 5.5
- Qwen2.5 7B (local) vs Claude Sonnet 5.5
- Qwen2.5 7B (local) vs DeepSeek V4.1 Flash (Venice)
- Qwen2.5 7B (local) vs Gemma 4 31B Instruct (Venice)
- Qwen2.5 7B (local) vs Grok 4.3 (Venice)
- Qwen2.5 7B (local) vs Llama 3.2 3B (local)
- Qwen2.5 7B (local) vs Mistral Small 3.2 24B Instruct (Venice)
- Qwen2.5 7B (local) vs Qwen3 Coder 480B Turbo (Venice)
Browse every run by Qwen2.5 7B (local) →
From published single-model runs only (all settings grouped). How we score.