bison-1 vs CCORE Web Coder 4B
Compare documented capabilities and access. A dash means a comparable figure is unavailable.
| Specification | ||
|---|---|---|
| Task | Decision scoring | Web code generation |
| Published capabilities | Post-trained from Gemma 4 31B IT for classification, routing and rubric scoring. | Fine-tuned from Qwen3.5-4B-Base to generate a standalone page or slide deck from a brief. |
| Model size | 31B parameters | 4B class |
| Access | Public weights and Baseten Model APIs as decision-models/baseten-bison-latest | Public 4.48 GB Q8_0 GGUF preview checkpoint |
| Result | Typed decisions and probabilities for supplied options | A self-contained React HTML document |
| Operating limits | Up to 131,072 tokens per question on Baseten Model APIs. | The publisher reports approximately 20,000 generated tokens for a typical whole document, including reasoning. |
| Billing terms | $0.10 USD per million input tokens on Baseten Model APIs. | Not published |
| Sources | Model weights · Additional documentation | Model weights |
| Inputs | Text, Images | Text |
| Context window | 131,072 tokens | Not published |
| Weights | Open weights · MIT and Apache-2.0 | Open weights · Apache-2.0 |
| Released | Nov 20, 2026 | Nov 20, 2026 |