NEXS Multi-SLERP Merges
A collection of four domain-expert model merges, one per architecture family, each produced by multi-SLERP in task-vector space with mergekit. Every merge is strictly intra-family — variants are only combined with other variants that share the same architecture and base model.
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.
For families whose variants had small vocabulary differences (a few added/removed special tokens),
tokenizer_source: base aligns every variant's embed_tokens/lm_head to the shared base
tokenizer before the merge, so all task vectors live in one embedding space.
Merged models
| Model | Architecture | Base | Merged variants |
|---|---|---|---|
| NEXS Llama-3.1-8B Multi-SLERP Merge | LlamaForCausalLM |
meta-llama/Llama-3.1-8B-Instruct | Llama-3.1-Hawkish-8B, Llama-3.1-8B-UltraMedical, calme-2.3-legalkit-8b, Llama-SafetyGuard-Content-Binary, Llama-3.1-8B-Instruct-multi-truth-judge |
| NEXS Qwen3-32B Multi-SLERP Merge | Qwen3ForCausalLM |
Qwen/Qwen3-32B | Light-IF-32B, MedGo, T-pro-it-2.0 |
| NEXS Deepseek-7B Multi-SLERP Merge | LlamaForCausalLM |
deepseek-ai/deepseek-llm-7b-base | deepseek-math-7b-instruct, deepseek-coder-7b-instruct-v1.5 |
| NEXS Qwen2.5-7B Multi-SLERP Merge | Qwen2ForCausalLM |
Qwen/Qwen2.5-7B | Qwen2.5-Coder-7B-Instruct, Qwen2.5-Math-7B |