fatalities003 is the latest version of the fatalities model. It has been in use since November 2025, generating monthly predictions for impending state-based conflict across the world up to three years in advance.
Important notice: In November 2025, we made new forecasts available from both the fatalities003 and fatalities002 models to allow for comparison. Both datasets are available through the VIEWS API as well as for download in the data section of this website. The data dashboard is however still limited to forecasts from the fatalities002 model. From December 2025 onwards, the fatalities002 model will be discontinued and replaced by fatalities003.
Predicted outcome #1
Number of fatalities
Point predictions for the number of fatalities per country-month and PRIO-GRID month.
Predicted outcome #2
Probability of conflict
Predicted probability of at least 25 battle-related deaths (BRDs) per country-month and at least 1 BRD per PRIO-GRID-month.
Predicted type(s) of violence
State-based conflict
Per UCDP definition: inter- or intrastate armed conflicts over government or territory, in which at least one of the warring parties are directly affiliated with a government of a state.
Country-level coverage
Global
The country level of analysis is based on the Gleditsch & Ward (1999) list of independent states, combined with the GIS dataset CShapes that specifies the geographic coverage of the included countries.
Sub-national coverage
Africa + Middle East (0.5°)
The sub-national level of analysis is derived from PRIO-GRID 2.0, a spatial grid structure of quadratic cells that jointly cover all areas of the world at a resolution of 0.5 x 0.5 decimal degrees, approximately 55×55 km around the equator.
Lead time
1-36 months
The model generates predictions for each month in a rolling 3-year window.
Update schedule
Monthly
The model generates new predictions each month, based on the most recently available input data.
Codebase
Open code
Source code and additional documentation of this model is available in a public GitHub repository.
How Are the Forecasts Generated?
What Data Informs the Model?
Input data (predictors)
The model is informed by nearly 200 open-source variables from data providers such as the Uppsala Conflict Data Program (UCDP), ACLED, PRIO-GRID, the World Bank, V-Dem, and FAO.
Browse the input data catalogue for the fatalities003 model
Model details
Delve deeper into the model specifications
The fatalities003 model is specified in an open-source code repository on GitHub: views-models. To consult the production model specifications, navigate to the ensembles called pink_ponyclub (the country-level ensemble) and skinny_love (the PRIO-GRID level ensemble).
How the Model Differs From Its Predecessor
Note that the list of source indicators informing the fatalities003 model, as well as the means in which the sub-models are implemented in the latter, differs from the predecessor model (fatalities002). The ensembling technique at the country level has also changed from a genetic algorithm to a simple unweighted average. Last, the surrogate model forecasts available from the fatalities002 model have now been discontinued. The main forecasts from the models, however, remain the same: both models produce forecasts for the number of fatalities from state-based armed conflict, in both logged and non-logged form. They both also produce dichotomous forecasts for the probability of observing at least 25 battle-related deaths per country-month and at least one battle-related death per PRIO-GRID cell and month.
Related Content
News, events, and publications related to this model
-
Posted on 30 Jan 2026
New Data Release: Forecasts for Jan 2026 – Dec 2028
The VIEWS forecasts for state-based armed conflict from January 2026 to December 2028 are now available. The dataset includes country-month forecasts and geo-referenced predictions for Africa and the Middle East, detailing expected direct deaths and probabilities.
Topics: fatalities003 -
Posted on 31 Dec 2025
New Data Release: Forecasts for Dec 2025 – Nov 2028
The latest VIEWS forecasts for state-based armed conflict are now available, covering the 36-month period from December 2025 through November 2028. The dataset includes global forecasts at the country-month level, along with geo-referenced forecasts at 0.5° spatial resolution for Africa and the Middle East. It captures the following indicators for state-based armed conflict: 📥 Download…
Topics: fatalities003 -
Posted on 12 Dec 2025
New Forecasts of the Deadliest Conflict Zones in 2026
Press release – original posted by PRIO Ukraine, Palestine/Israel, Sudan, Pakistan and Nigeria are projected to see the highest battle-related death tolls in 2026, according to a world-leading AI-driven conflict forecasting system. The Violence and Impacts Early-Warning System (VIEWS) is an award-winning conflict prediction model developed by the Peace Research Institute Oslo (PRIO) and Uppsala University, that…
-
Posted on 28 Nov 2025
New Conflict Forecasting Model Released: Introducing ‘fatalities003’
We’re thrilled to announce the release of the new fatalities003 conflict prediction model – running on an entirely new production pipeline. What’s New? Building on its predecessor (fatalities002), the new model retains the same sub-model structure, with each sub-model corresponding to a counterpart in the previous version. However, several key changes have been made: Forecasts…
Topics: fatalities003
