This model is the quantized version of NexusFlow's NexusRaven V-2. The quantization technique used is Activation-Aware Weight Quantization. The model is suitable for high-degree of Function Calling. The functions may be Simple Functions, Compound Functions or Nested Functions. The model hasn't been fine-tuned yet.

Model creator: Nexusflow

Original model: NexusRaven V2 13B

Downloads last month
19
Safetensors
Model size
2.03B params
Tensor type
I32
·
FP16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train NaiveAttention/NexusRaven-V2-13B-awq