What is the difference between machine learning model and ML algorithm?

I will answer this with an example. Linear regression algorithm is a technique to fit points to a line y = m x+c. Now after fitting, you get for example, y = 10 x + 4. This a model. A model is something to which when you give an input, gives an output. In ML, any 'object' created after training from an ML algorithm is a model. For example, SVM model, Random forests model, etc.

The “ML model” is the output generated when you train your “machine learning algorithm” with your training data-set.For example, if I train my Decision Tree algorithm with a structured training data-set for say, anomaly detection in a network to identify malicious packets, it will generate a model which would take in an input, preferably in real time, and generate a result set corresponding to each packet. The training set would include normal and malicious packets. Each packet’s information (in this case, packet size, protocol, duration and so on) needs to be resolved as per the choice of features (or attributes) and algorithm.The important thing is that the model generated has to take in a predefined input structure.Example: 1, 23, 454, 55, 7It cannot take in an input like: 5, 67, 89The number of features for your test data to be fed into your ML model has to be equal to the number of features used during training, in case of supervised learning.The ML model can also be dynamic, in the sense that it can train itself on the fly upon encountering new data. This can be explained in the case of K-means clustering algorithm. After creation of the initial model, the data would be classified into clusters and any new encounters would make the boundaries of separation clearer, thus improving the model’s ability for classification (using layman’s terms here).

I don't think there is any real difference. But if I had to pick then I'd say the model is what it is after learning and the algorithm is the method of learning. For example, after training with lots of rgb values using the SVM algorithm we have created a new skin model.