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 | Orion Coder 7B open-lab/orion-coder-7b |
|---|---|---|
| Openness | Open weights | Open weights |
| Commercial use | Permitted | Permitted |
| Parameters | 8.1B | 7.2B |
| Context length | 66K tokens | 33K tokens |
| Estimated memory | 6.8 GB · Q5_K_M | 5.6 GB · Q4_K_M |
| Runtimes | Ollama, llama.cpp, Transformers, vLLM | Ollama, llama.cpp, MLX, Transformers, vLLM |
| Tasks | Reasoning, Mathematics, Question answering | Coding, Code review, Tool use |
| Verified Runs | 14 | 12 |
| Best evidence | Standardized evaluation | Standardized evaluation |
| Task score | 68.7 Reasoning suite | 74.2 Coding suite |
| Speed | 56.9 tok/s | 42.8 tok/s |
| Peak memory | 8.3 GB | 6.1 GB |
| Run hardware | RTX 4090 · 24 GB | Apple M3 Max · 64 GB |
An open-weight reasoning model designed for careful, inspectable local evaluations.
A compact code-generation model tuned for local development workflows.
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.
