google-deepmind / google-deepmind/weathernext

GenCast Running in BFloat16 Precision Available?

Open
#114 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
7.7k
Forks
986
PR merge metrics
No merged PRs in 30d

Description

Does GenCast model have any chance that running in bfloat16 floating point precision like GraphCast?

Thanks!

Contributor guide

Open the contributing guide

Research direction

Start by reviewing the GenCast implementation and the existing GraphCast bfloat16 support referenced in the issue. Determine whether bfloat16 execution is supported and what code, documentation, or tests would need to change; the work is complete when the supported behavior and validation are clearly established.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.