Language model · The CALLAII language model prepared to run on the organization's own server.
CALLAII Edge v1In training
CALLAII Edge is currently in training. The goal is to consolidate CALLAII's legal and call center capabilities into a model that runs on the organization's own server. The graph below is updated regularly during training.
What improved in this version
Goal: Identify the competent and authorized court, litigation prerequisites, and deadlines for a legal question
Goal: Introduce itself as CALLAII under all circumstances
Results will be published here after training is complete, using the same exam taken before training
Training and measurements
0,49Epoch
0,6186Final validation loss
Status
In training
Training data
45,847 samples
Validation set
200 samples never seen during training
Training context
4,096 tokens
Training date
September 2026
Training and validation loss
Training lossValidation loss
Each point is the average of 3 consecutive records.
Show values as a table
Training loss
0.01
5.0917
0.03
0.9474
0.05
0.7875
0.07
0.7862
0.09
0.7166
0.11
0.7118
0.13
0.6670
0.15
0.6756
0.17
0.6324
0.19
0.6646
0.22
0.6551
0.24
0.6583
0.26
0.6638
0.28
0.6530
0.30
0.6599
0.32
0.6420
0.34
0.6337
0.36
0.6260
0.38
0.6009
0.40
0.6391
0.42
0.6311
0.45
0.6200
0.47
0.6200
0.49
0.6187
0.49
0.6022
Validation loss
0.07
0.7438
0.35
0.6291
0.42
0.6186
Customer data is never used to train the models.
This sheet was generated on 27.09.2026. See callmenta.com for the current version.
Versions and improvements
CALLAII®
by CALLMENTA