# AIFS **URL:** https://forum.ecmwf.int/c/ifs-aifs-and-openifs/aifs/43.md [Latest](https://forum.ecmwf.int/latest.md) · [Categories](https://forum.ecmwf.int/categories.md) · [Tags](https://forum.ecmwf.int/tags.md) --- ## [About the AIFS category](https://forum.ecmwf.int/t/about-the-aifs-category/11987)
**Author:** [@Michela](https://forum.ecmwf.int/u/Michela)\ **Replies:** 0\ **Last updated:** [20 March 2025 17:41 UTC](https://forum.ecmwf.int/t/about-the-aifs-category/11987 "2025-03-20T17:41:51Z")
Welcome to the AIFS forum! This is your hub for asking questions, and engaging with fellow users of the AIFS model. Be sure to check the instructions below to get started! To learn more about AIFS implementation, see th… --- ## [AIFS static fields timestamp](https://forum.ecmwf.int/t/aifs-static-fields-timestamp/15294)
**Author:** [@Aaron\_Spring](https://forum.ecmwf.int/u/Aaron_Spring)\ **Replies:** 0\ **Last updated:** [1 September 2026 14:59 UTC](https://forum.ecmwf.int/t/aifs-static-fields-timestamp/15294 "2026-09-01T14:59:27Z")
I’m trying to reproduce AIFS forecasts. I’m wondering which timestamp to download for the static fields. In particular I try to run AIFSv1.1 which requires static variables lsm, slor, sdor, and z\_sfc, essentially all ab… --- ## [AIFS V3 Fields and Levels?](https://forum.ecmwf.int/t/aifs-v3-fields-and-levels/15279)
**Author:** [@Jamie\_Ray](https://forum.ecmwf.int/u/Jamie_Ray)\ **Replies:** 0\ **Last updated:** [24 August 2026 12:34 UTC](https://forum.ecmwf.int/t/aifs-v3-fields-and-levels/15279 "2026-08-24T12:34:24Z")
I’m wondering if the fields / levels have been decided, and in that case what they are (and the diagnostic vs. prognostic breakdown). Thanks! --- ## [AIFS Single v2 and AIFS ENS v2](https://forum.ecmwf.int/t/aifs-single-v2-and-aifs-ens-v2/14907)
**Author:** [@hyy](https://forum.ecmwf.int/u/hyy)\ **Replies:** 2\ **Last updated:** [14 May 2026 10:51 UTC](https://forum.ecmwf.int/t/aifs-single-v2-and-aifs-ens-v2/14907 "2026-05-14T10:51:39Z")
Hello, while visualizing the test data for AIFS single v2 on the OpenData webpage, we unexpectedly discovered that the new AIFS model’s 06z and 12z showed abnormal differences in the simulation of the Northwest Pacific i… --- ## [503 Slowdown errors for requests to s3://ecmwf-forecasts](https://forum.ecmwf.int/t/503-slowdown-errors-for-requests-to-s3-ecmwf-forecasts/14345)
**Author:** [@Evan\_Schechter](https://forum.ecmwf.int/u/Evan_Schechter)\ **Replies:** 6\ **Last updated:** [12 May 2026 10:03 UTC](https://forum.ecmwf.int/t/503-slowdown-errors-for-requests-to-s3-ecmwf-forecasts/14345 "2026-05-12T10:03:01Z")
Hi all, Over the past few days, I’ve been noticing frequent 503 errors when requesting AIFS data from the s3://ecmwf-forecasts S3 bucket. aws s3 cp --no-sign-request s3://ecmwf-forecasts/20251101/00z/aifs-ens/0p25/enfo… --- ## [Inference with AIFS-ENS v.1 on S2S Timescales](https://forum.ecmwf.int/t/inference-with-aifs-ens-v-1-on-s2s-timescales/14832)
**Author:** [@Aodhan\_Sweeney](https://forum.ecmwf.int/u/Aodhan_Sweeney)\ **Replies:** 1\ **Last updated:** [29 April 2026 15:50 UTC](https://forum.ecmwf.int/t/inference-with-aifs-ens-v-1-on-s2s-timescales/14832 "2026-04-29T15:50:54Z")
Hello! Thanks for making the AIFS-ENS v1.0 available! I was quite impressed to see the skill of the AIFS-ENS v1.0 compared to the S2S runs of IFS shown in the recent AIFS-CRPS paper ( AIFS-CRPS: Ensemble forecasting usi… --- ## [AIFS ENS TC tracks](https://forum.ecmwf.int/t/aifs-ens-tc-tracks/13769)
**Author:** [@Peter\_Jermey](https://forum.ecmwf.int/u/Peter_Jermey)\ **Replies:** 8\ **Last updated:** [27 April 2026 08:25 UTC](https://forum.ecmwf.int/t/aifs-ens-tc-tracks/13769 "2026-04-27T08:25:34Z")
Hi Is there a plan to produce TC track files for the AIFS ensemble similar to s3://ecmwf-forecasts/20250706/12z/aifs-single/0p25/oper/20250706120000-240h-oper-tf.bufr? Many thanks --- ## [Reading and regridding from cloud-based datasets](https://forum.ecmwf.int/t/reading-and-regridding-from-cloud-based-datasets/14659)
**Author:** [@Julian\_Schmitt](https://forum.ecmwf.int/u/Julian_Schmitt)\ **Replies:** 2\ **Last updated:** [6 April 2026 20:45 UTC](https://forum.ecmwf.int/t/reading-and-regridding-from-cloud-based-datasets/14659 "2026-04-06T20:45:49Z")
I am trying to build a xarray-zarr cloud-based dataset with ERA5 for training a 1° model and am finding that regridding step of the creation is extremely slow. My understanding is that the anemoi dataset .zarr lives in t… --- ## [Does switching variables from prognostic to forcing necessitate training from scratch?](https://forum.ecmwf.int/t/does-switching-variables-from-prognostic-to-forcing-necessitate-training-from-scratch/14723)
**Author:** [@Julian\_Schmitt](https://forum.ecmwf.int/u/Julian_Schmitt)\ **Replies:** 2\ **Last updated:** [6 April 2026 20:36 UTC](https://forum.ecmwf.int/t/does-switching-variables-from-prognostic-to-forcing-necessitate-training-from-scratch/14723 "2026-04-06T20:36:38Z")
We are experimenting with changing some variables from prognostic to forcing. We are wondering if in changing the model architecture like this we are still able to fine-tune a current AIFS checkpoint or do we necessarily… --- ## [AIFS Hindcasts/Re-forecasts](https://forum.ecmwf.int/t/aifs-hindcasts-re-forecasts/14507)
**Author:** [@Brian\_Etherton1](https://forum.ecmwf.int/u/Brian_Etherton1)\ **Replies:** 1\ **Last updated:** [31 March 2026 11:57 UTC](https://forum.ecmwf.int/t/aifs-hindcasts-re-forecasts/14507 "2026-03-31T11:57:42Z")
Hello there one and all, I was curious if there had been any desire to run hindcasts for the AIFS ensemble for cases prior to 2025. Perhaps go back a few years? I ask because such hindcasts (perhaps a better term is ‘r… --- ## [Soil temperature representation in AIFS v1 and precipitation results](https://forum.ecmwf.int/t/soil-temperature-representation-in-aifs-v1-and-precipitation-results/14004)
**Author:** [@slava.kohut](https://forum.ecmwf.int/u/slava.kohut)\ **Replies:** 1\ **Last updated:** [18 March 2026 14:51 UTC](https://forum.ecmwf.int/t/soil-temperature-representation-in-aifs-v1-and-precipitation-results/14004 "2026-03-18T14:51:03Z")
The AIFS v1 model is trained on two datasets: ERA5 and IFS (operational analysis). The latter is used for fine tuning. The way soil temperature is represented in these datasets is different: there is a mask for oceans in… --- ## [Snowfall variable available on "ERA5 hourly data on single levels from 1940 to present" but not on "Complete ERA5 global atmospheric reanalysis"?](https://forum.ecmwf.int/t/snowfall-variable-available-on-era5-hourly-data-on-single-levels-from-1940-to-present-but-not-on-complete-era5-global-atmospheric-reanalysis/14774)
**Author:** [@Soulivanh\_Thao](https://forum.ecmwf.int/u/Soulivanh_Thao)\ **Replies:** 2\ **Last updated:** [10 March 2026 12:28 UTC](https://forum.ecmwf.int/t/snowfall-variable-available-on-era5-hourly-data-on-single-levels-from-1940-to-present-but-not-on-complete-era5-global-atmospheric-reanalysis/14774 "2026-03-10T12:28:50Z")
Hello, I trying to learn how to use anemoi by trying to retraining AIFS 1.10 with the procedure noted in the huggingface notebook. To do that, I am trying to download with (anemoi-datasets or the CDS API) the relevant … --- ## [Seeking architecture and training configuration advice for 1° model](https://forum.ecmwf.int/t/seeking-architecture-and-training-configuration-advice-for-1-model/14722)
**Author:** [@Julian\_Schmitt](https://forum.ecmwf.int/u/Julian_Schmitt)\ **Replies:** 0\ **Last updated:** [12 February 2026 23:27 UTC](https://forum.ecmwf.int/t/seeking-architecture-and-training-configuration-advice-for-1-model/14722 "2026-02-12T23:27:51Z")
We are training a 1° model (O96) from scratch on the provided 40 year ERA5 dataset using a slightly modified default configuration for MSE training. We have modified the following configuration arguments and trained an i… --- ## [Difficulties running inference with AIFS v1.1](https://forum.ecmwf.int/t/difficulties-running-inference-with-aifs-v1-1/14533)
**Author:** [@Tom\_Jerry](https://forum.ecmwf.int/u/Tom_Jerry)\ **Replies:** 0\ **Last updated:** [17 December 2025 15:07 UTC](https://forum.ecmwf.int/t/difficulties-running-inference-with-aifs-v1-1/14533 "2025-12-17T15:07:42Z")
Hello everyone, I’m trying to run inference with AIFS v1.1, following the approach in run\_AIFS\_v1.1.ipynb. I’ve tested two different setups: Running a forecast based on IFS data with minimal modifications to the note… --- ## [Issue with AIFS](https://forum.ecmwf.int/t/issue-with-aifs/11593)
**Author:** [@tan\_guihong](https://forum.ecmwf.int/u/tan_guihong)\ **Replies:** 3\ **Last updated:** [8 August 2025 08:26 UTC](https://forum.ecmwf.int/t/issue-with-aifs/11593 "2025-08-08T08:26:33Z")
I found that AIFS data has no data since 2025-02-24 06 UTC. Will it stop updating in the future --- ## [AIFS ensemble model](https://forum.ecmwf.int/t/aifs-ensemble-model/13734)
**Author:** [@Antonio\_Bruno](https://forum.ecmwf.int/u/Antonio_Bruno)\ **Replies:** 4\ **Last updated:** [8 July 2025 12:32 UTC](https://forum.ecmwf.int/t/aifs-ensemble-model/13734 "2025-07-08T12:32:39Z")
Hi there, on regarding the new released notebook to run AIFS\_ens on Huggingface, I want to use it to run an ensemble forecast but I understood that the notebook is made to run only an element of the ensamble with the or… --- ## [AIFS ENS 2024 history](https://forum.ecmwf.int/t/aifs-ens-2024-history/13590)
**Author:** [@Hugh\_Baker](https://forum.ecmwf.int/u/Hugh_Baker)\ **Replies:** 3\ **Last updated:** [24 June 2025 10:32 UTC](https://forum.ecmwf.int/t/aifs-ens-2024-history/13590 "2025-06-24T10:32:47Z")
Hi there, On reading the AIFS ENS CRPS paper it seems as if the figures use N320 data \[0 and 12z\] from 2024 Feb 1st onwards. Will this data be made available as well as the planned backhistory from Dec 2024 stated in th… --- ## [AIFS expanded products](https://forum.ecmwf.int/t/aifs-expanded-products/13606)
**Author:** [@Hugh\_Baker](https://forum.ecmwf.int/u/Hugh_Baker)\ **Replies:** 1\ **Last updated:** [24 June 2025 10:24 UTC](https://forum.ecmwf.int/t/aifs-expanded-products/13606 "2025-06-24T10:24:52Z")
Hello, Following the AIFS-ENS seminar and the request for users for requests for expanded products, it would be great for us to see tmax, tmin and fdir in the outputs if possible please! Many thanks, Hugh --- ## [AIFS Historical Archive and Hindcasts?](https://forum.ecmwf.int/t/aifs-historical-archive-and-hindcasts/12589)
**Author:** [@Arbol\_Data\_Team](https://forum.ecmwf.int/u/Arbol_Data_Team)\ **Replies:** 3\ **Last updated:** [3 June 2025 19:01 UTC](https://forum.ecmwf.int/t/aifs-historical-archive-and-hindcasts/12589 "2025-06-03T19:01:58Z")
Hi, I’m interested in testing the performance of AIFS against alternative datasets for some applications. This requires historical data and ideally hindcasts showing how the model would have performed previously. Howev… --- ## [How to output the forecast hour grib data](https://forum.ecmwf.int/t/how-to-output-the-forecast-hour-grib-data/12724)
**Author:** [@RAN\_HUANG](https://forum.ecmwf.int/u/RAN_HUANG)\ **Replies:** 0\ **Last updated:** [22 April 2025 11:40 UTC](https://forum.ecmwf.int/t/how-to-output-the-forecast-hour-grib-data/12724 "2025-04-22T11:40:19Z")
Hi, We are running aifs-v1.0 on our local machine and encountering an issue with the forecast output time steps. Currently, when running a forecast initialized at a specific time (e.g., 2025041700z), the first output in…