Cygnis-A3-V2-Light

Version allégée et optimisée de Cygnis-A3 réalisée avec mergekit.

🧩 Modèles utilisés (1)

📜 Configuration

base_model: Qwen/Qwen2.5-7B-Instruct

models:
  - model: Qwen/Qwen2.5-Coder-7B-Instruct
  - model: Qwen/Qwen2.5-Math-7B-Instruct

parameters:
  strength: 0.85
  retention: 0.65
  novelty: 0.30
  stability: 0.75

dtype: float32
out_dtype: bfloat16

tokenizer:
  source: Qwen/Qwen2.5-7B-Instruct```
Downloads last month
47
Safetensors
Model size
8B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support