Modelverse keeps model facts separate from Run evidence and warns whenever benchmark conditions are not equivalent.
These models do not currently have leading Runs under identical evaluation conditions. Values are shown with their provenance and should not be treated as a universal leaderboard.
| Attribute | Aurora Reason 8B open-lab/aurora-reason-8b | Nova Instruct 12B community/nova-instruct-12b |
|---|---|---|
| Openness | Open weights | Open weights |
| Commercial use | Permitted | Permitted |
| Parameters | 8.1B | 12B |
| Context length | 66K tokens | 33K tokens |
| Estimated memory | 6.8 GB · Q5_K_M | 8.2 GB · Q4_K_M |
| Runtimes | Ollama, llama.cpp, Transformers, vLLM | Ollama, llama.cpp, Transformers, vLLM |
| Tasks | Reasoning, Mathematics, Question answering | Chat, Summarization, Reasoning, Translation |
| Verified Runs | 14 | 9 |
| Best evidence | Standardized evaluation | Machine-captured |
| Task score | 68.7 Reasoning suite | 71.3 Multilingual instruction suite |
| Speed | 56.9 tok/s | 118.4 tok/s |
| Peak memory | 8.3 GB | 18.8 GB |
| Run hardware | RTX 4090 · 24 GB | RTX 3090 · 24 GB |
An open-weight reasoning model designed for careful, inspectable local evaluations.
A general-purpose multilingual instruction model with broad runtime support.
First: eliminate models that fail the license, task, modality or runtime requirement.
Second: confirm fit on equivalent hardware or preserve a generous estimated memory margin.
Third: compare only Runs sharing an evaluation suite, revision and broadly equivalent configuration.
