Updated model card

README.md

@@ -1,11 +1,8 @@
 ---
 license: apache-2.0
 ---
----
-license: apache-2.0
----
 # FlowState
-[Paper](https://www.arxiv.org/abs/2508.05287) | [HuggingFace Model Card](https://huggingface.co/ibm-granite/granite-timeseries-flowstate-r1) | [GitHub Model Code](https://github.com/ibm-granite/granite-tsfm/tree/main/tsfm_public/models/flowstate)
+[Paper](https://www.arxiv.org/abs/2508.05287) | [HuggingFace Model Card](https://huggingface.co/ibm-research/flowstate) | [GitHub Model Code](https://github.com/ibm-granite/granite-tsfm/tree/main/tsfm_public/models/flowstate)
 
 ![Illustration](figs/FlowState.png)
 FlowState is the first time-scale adjustable Time Series Foundation Model (TSFM), open-sourced by IBM Research.
@@ -35,7 +32,8 @@ FlowState can be used to make predictions as follows:
 from tsfm_public import FlowStateForPrediction
 import torch
 device= 'cuda'
-predictor = FlowStateForPrediction.from_pretrained("ibm-granite/granite-timeseries-flowstate-r1").to(device)
+# Download FlowState Research checkpoint (non-commercial use):
+predictor = FlowStateForPrediction.from_pretrained("ibm-research/flowstate").to(device)
 time_series = torch.randn((2048, 32, 1), device=device) # context, batch, n_ch
 forecast = predictor(time_series, scale_factor=0.25, prediction_length=960, batch_first=False)
 print(forecast.prediction_outputs.shape) # torch.Size([32, 9, 48, 1]) (batch, quantiles, forecast_length, n_ch)