Bonsai Models by PrismML
Last refreshed 2026-10-22. Next refresh: weekly.
Details
Capabilities
About
Bonsai is PrismML's family of 1-bit and ternary open-weight language models for local / edge deployment (docs: 'Bonsai is a family of 1-bit and ternary language models from PrismML'). Line includes Ternary Bonsai 2 27B (2026-09-17; based on Qwen3.8 27B), prior Bonsai 27B generation (2026-07-14; based on Qwen3.6 27B — Ternary + 1-bit variants), smaller Bonsai 8B/4B/1.7B, and Bonsai Image. Ternary Bonsai 2 27B is the current flagship, with prior Ternary Bonsai 27B as a sibling under the same family.
Decision facts
- Best fit
- agentson deviceimage generation
- Capability starting point
- Ternary Bonsai 2 2B with 256K context and reasoning, JSON / Tool use, and multimodal inputs
- Lowest tracked input
- Not tracked
- Closest related family
- Agents-A1
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs on device, 256K context, and 1.9B parameters.
Use when the workload needs on device, 256K context, and 4.2B parameters.
Use when the workload needs on device, 256K context, and 8.9B parameters.
Use when the workload needs on device, 256K context, and 27.4B parameters.
Use when the workload needs on device, 256K context, and 27B parameters.
Use when the workload needs on device, 256K context, and 27.4B parameters.
Use when the workload needs image generation, 4B parameters, and multimodal inputs.
Use when the workload needs image generation, 4B parameters, and multimodal inputs.
Use when the workload needs on device, 32K context, and 1.7B parameters.
Use when the workload needs on device, 32K context, and 4B parameters.
Use when the workload needs on device, 64K context, and 8.2B parameters.
Use when the workload needs on device, 32K context, and 1.7B parameters.
Use when the workload needs on device, 32K context, and 4B parameters.
Use when the workload needs on device, 64K context, and 8.2B parameters.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Ternary Bonsai 2 2B | Use when the workload needs on device, 256K context, and 1.9B parameters. | 2026-10 | on device256K context1.9B parameters | Current |
| Ternary Bonsai 2 4B | Use when the workload needs on device, 256K context, and 4.2B parameters. | 2026-10 | on device256K context4.2B parameters | Current |
| Ternary Bonsai 2 9B | Use when the workload needs on device, 256K context, and 8.9B parameters. | 2026-10 | on device256K context8.9B parameters | Current |
| Ternary Bonsai 2 27B | Use when the workload needs on device, 256K context, and 27.4B parameters. | 2026-09 | on device256K context27.4B parameters | Current |
| Ternary Bonsai 27B | Use when the workload needs on device, 256K context, and 27B parameters. | 2026-07 | on device256K context27B parameters | Current |
| Bonsai 27B | Use when the workload needs on device, 256K context, and 27.4B parameters. | 2026-07 | on device256K context27.4B parameters | Current |
| Bonsai Image 4B | Use when the workload needs image generation, 4B parameters, and multimodal inputs. | 2026-05 | image generation4B parametersmultimodal inputs | Current |
| Ternary Bonsai Image 4B | Use when the workload needs image generation, 4B parameters, and multimodal inputs. | 2026-05 | image generation4B parametersmultimodal inputs | Current |
| Ternary Bonsai 1.7B | Use when the workload needs on device, 32K context, and 1.7B parameters. | 2026-04 | on device32K context1.7B parameters | Current |
| Ternary Bonsai 4B | Use when the workload needs on device, 32K context, and 4B parameters. | 2026-04 | on device32K context4B parameters | Current |
| Ternary Bonsai 8B | Use when the workload needs on device, 64K context, and 8.2B parameters. | 2026-04 | on device64K context8.2B parameters | Current |
| Bonsai 1.7B | Use when the workload needs on device, 32K context, and 1.7B parameters. | 2026-03 | on device32K context1.7B parameters | Current |
| Bonsai 4B | Use when the workload needs on device, 32K context, and 4B parameters. | 2026-03 | on device32K context4B parameters | Current |
| Bonsai 8B | Use when the workload needs on device, 64K context, and 8.2B parameters. | 2026-03 | on device64K context8.2B parameters | Current |
Release Timeline
6 release groupsSpecifications(14 models)
| Model | Released | Context | Parameters | Vision | Multimodal | Reasoning | JSON / Tool use |
|---|---|---|---|---|---|---|---|
| Ternary Bonsai 2 2B | 2026-10 | 256K | 1.88B | Yes | Yes | Yes | Yes |
| Ternary Bonsai 2 4B | 2026-10 | 256K | 4.21B | Yes | Yes | Yes | Yes |
| Ternary Bonsai 2 9B | 2026-10 | 256K | 8.95B | Yes | Yes | Yes | Yes |
| Ternary Bonsai 2 27B | 2026-09 | 256K | 27.4B | Yes | Yes | Yes | Yes |
| Ternary Bonsai 27B | 2026-07 | 256K | 27B | Yes | Yes | Yes | Yes |
| Bonsai 27B | 2026-07 | 256K | 27.36B | Yes | Yes | Yes | Yes |
| Bonsai Image 4B | 2026-05 | — | 4B | Yes | Yes | No | No |
| Ternary Bonsai Image 4B | 2026-05 | — | 4B | Yes | Yes | No | No |
| Ternary Bonsai 1.7B | 2026-04 | 32K | 1.72B | No | No | No | No |
| Ternary Bonsai 4B | 2026-04 | 32K | 4.0B | No | No | No | No |
| Ternary Bonsai 8B | 2026-04 | 64K | 8.19B | No | No | No | No |
| Bonsai 1.7B | 2026-03 | 32K | 1.7B | No | No | No | No |
| Bonsai 4B | 2026-03 | 32K | 4.0B | No | No | No | No |
| Bonsai 8B | 2026-03 | 64K | 8.19B | No | No | No | No |
Available From(1 provider)
Models(14)
Ternary Bonsai 2 2B
Ternary Bonsai 2 4B
Ternary Bonsai 2 9B
Ternary Bonsai 2 27B
Ternary Bonsai 27B
Bonsai 27B
Bonsai Image 4B
Ternary Bonsai Image 4B
Ternary Bonsai 1.7B
Ternary Bonsai 4B
Ternary Bonsai 8B
Bonsai 1.7B
Bonsai 4B
Bonsai 8B