Update README.md

README.md

@@ -45,6 +45,16 @@ The training process consisted of three distinct stages:
 - Utilized a distillation dataset of 900K samples
 - Implemented VariableRankPEFT to enhance training efficiency
 
+## How to Use
+- install RWKV-Infer(see how to install) https://github.com/OpenMOSE/RWKV-Infer
+- loadmodel(choose fp16 or fp6 or fp5 (dont choose FP8))
+- need 19GB VRAM in FP16, 12GB VRAM in FP6
+```
+curl http://127.0.0.1:9000/loadmodel -X POST -H "Content-Type: application/json" -d '{"model_filename":"models/rwkv-phi4-stage2.pth","model_viewname":"PRWKV7-Phi-4 Preview 0.1","model_strategy":"fp5","template":"phi4"}'
+
+```
+3. you can use this model via OpenAI CompatibleAPI http://127.0.0.1:9000/v1 and set modelname "PRWKV7-Phi-4 Preview 0.1"
+
 ## Training Infrastructure
 - Hardware: Single AMD MI300X GPU
 - Training Duration: 3 days(Stage1,2)
@@ -59,6 +69,10 @@ This work was made possible through the contributions of:
 - RecursalAI
 - RWKV-Red-Team
 
+- https://github.com/OpenMOSE/RWKVInside
+- https://github.com/OpenMOSE/RWKV-LM-RLHF
+- https://github.com/OpenMOSE/RWKV-Infer
+
 ## Limitations
 
 This is trained Stage2. Stage3 is on going :)