Version bump and slight README update

Author: baldikacti

README.md

@@ -4,11 +4,13 @@ This is a convinience [Alphafold3](https://github.com/google-deepmind/alphafold3
 
 - Splits MSA (CPU-heavy) and structure prediction (GPU-heavy) steps and submits jobs accordingly
 
+- GPU jobs will be batched for efficiency. 20 predictions will be run on each GPU.
+
 - Parallel execution and checkpointing
 
 - Smart resource usage. Failed jobs will be automatically submitted again with higher resources including better GPUs when neccessary.
 
-- The pipeline will use GPUs with GPU compute capability > 8 to prevent problems. 
+- The pipeline will use GPUs with GPU compute capability > 8 to prevent problems.
 
 # Usage
 
@@ -43,7 +45,7 @@ module load nextflow/24.10.3
 
 export NXF_OPTS="-Xms1G -Xmx8G"
 
-nextflow run UnityHPC/alphafold3-batch \
+nextflow run UnityHPC/alphafold3-batch -r v0.2.0 \
     --json_dir /path/to/json_dir \
     --output_dir /path/to/results \
     --model_dir /path/to/model/dir \