Ternary Bonsai 2 27B
PrismML · Lanzado sept 2026
Ternary Bonsai 2 27B squeezes Qwen3.8 27B into about 5.9GB at 1.76 effective bits, so a 262K-context text-and-image model fits on a Mac or a single consumer GPU under Apache 2.0. Conditional because the quality figures are PrismML's own, its blog and model card report different aggregate scores, and the files need PrismML's custom kernels: stock llama.cpp will not run them. Right for local-inference users short on memory.
¿Es adecuada para ti?
Bueno para
- Running a 27B-class model in about 6GB: the ternary GGUF is 5.95GB against a 53.8GB FP16 reference, plus a 0.63GB projector for image input
- Apple silicon Macs: PrismML reports 46.8 tokens per second on an M5 Max through MLX
- Long-context local work with a 262K-token window
- Commercial use and redistribution under Apache 2.0
No recomendado para
- Standard local tooling: the GGUF files need PrismML's llama.cpp fork, and stock llama.cpp will not run them
- Maximum quality: PrismML's own figures put it at 98.2% of full-precision Qwen3.8 27B
- Decisions that need independent evaluations: all published scores are PrismML's
Precios
Entrada
Free (open weights)
Salida
Free (open weights)
Contexto
262K ctx; self-hosted; PrismML kernels required
Pruebas de rendimiento
Aún no hay cambios de veredicto
El reloj corre desde el primer día: los cambios aparecerán aquí a medida que evolucione nuestro veredicto.
Fuentes
Registro de verificación
- Perfil· Sin cambios
Agente automatizado
- Precios· Sin cambios
Agente automatizado
- Perfil· Sin cambios
Agente automatizado
- Precios· Sin cambios
Agente automatizado
- Perfil· Sin cambios
Agente automatizado
- Precios· Sin cambios
Agente automatizado