Method
Multi-SLERP (multislerp) performs barycentric spherical interpolation on a hypersphere for more than two models: it projects the models into the tangent space at their weighted Euclidean mean, interpolates, and projects back. Here it is run in task-vector space — each source's delta from the shared base model is computed, the deltas are spherically averaged with equal weight (normalize_weights: true, eps: 1e-8), and the result is added back to the base. Merging was done with mergekit.
Variants had minor vocab differences (151936 vs 151668); tokenizer_source: base reconciled all embeddings to the base tokenizer.
Sources
Base model (task-vector reference): Qwen/Qwen3-32B
Merged variants (equal weight 1.0 each):
mergekit config
merge_method: multislerp
base_model: Qwen/Qwen3-32B
tokenizer_source: base
dtype: float32
out_dtype: bfloat16
parameters:
normalize_weights: true
eps: 1.0e-8
models:
- model: qihoo360/Light-IF-32B
parameters: {weight: 1.0}
- model: OpenMedZoo/MedGo
parameters: {weight: 1.0}
- model: t-tech/T-pro-it-2.0
parameters: {weight: 1.0}