{"apiVersion":"v1","taxonomyVersion":1,"generatedAt":"2026-10-29T14:02:38.106Z","data":{"recordId":"wsa_u9tkz9tipder","score":0.933,"confidence":"high","capped":false,"capReason":null,"sources":[{"n":1,"url":"https://mistral.ai/news/timestral/","publisher":"Mistral","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-29T00:08:40.326Z","detectedPublishedAt":"2026-07-15T00:00:00.000Z","rights":"original"},{"n":2,"url":"https://huggingface.co/mistralai/Timestral-209M","publisher":"Hugging Face","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-29T00:08:40.326Z","detectedPublishedAt":"2026-10-14T08:52:36.000Z","rights":"original"},{"n":3,"url":"https://huggingface.co/mistralai/Timestral-5M","publisher":"Hugging Face","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-29T00:08:40.326Z","detectedPublishedAt":"2026-10-14T13:41:13.000Z","rights":"original"},{"n":4,"url":"https://huggingface.co/mistralai/Timestral-28M","publisher":"Hugging Face","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-15T16:31:12.197Z","detectedPublishedAt":"2026-10-14T17:50:53.000Z","rights":"original"},{"n":5,"url":"https://huggingface.co/mistralai/Timestral-123M","publisher":"Hugging Face","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-29T00:08:40.326Z","detectedPublishedAt":"2026-10-14T17:51:04.000Z","rights":"original"},{"n":6,"url":"https://github.com/mistralai/timestral","publisher":"GitHub new repositories","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-29T00:08:40.326Z","detectedPublishedAt":"2026-10-15T10:20:18.000Z","rights":"original"}],"claims":[{"id":"e9ac0749-a267-4534-a470-928bd5d0127e","kind":"availability","statement":"Timestral-5M weights are available from the Mistral AI organization on the Hugging Face Hub.","citations":[{"source":6,"supportProbability":0.92,"supports":true,"quote":{"text":"The weights are on the Hugging Face Hub, in the Mistral AI organization:","rights":"original"}}]},{"id":"a8a81351-1d49-4d17-a10a-d28dfcfe1acf","kind":"capability","statement":"The 5M model takes 17 ms to forecast one day ahead from three weeks of 15-minute history on a standard server CPU with 8 threads.","citations":[{"source":1,"supportProbability":0.98,"supports":true,"quote":{"text":"On a standard server CPU with 8 threads, forecasting one day ahead from three weeks of 15-minute history takes 17 ms with the 5M model and 113 ms with the 209M model.","rights":"original"}}]},{"id":"3734db58-b7de-4bb5-95b2-0ab21680c8ce","kind":"deployment","statement":"The Timestral pipeline can be deployed on-prem, in a VPC or in an air-gapped environment.","citations":[{"source":1,"supportProbability":0.88,"supports":true,"quote":{"text":"The full pipeline can be deployed on our clients’ hardware, including air-gapped environments.","rights":"original"}}]},{"id":"d6d99b2e-cb70-47ed-8446-067feb62f3dc","kind":"capability","statement":"Timestral-5M has 5M parameters.","citations":[{"source":3,"supportProbability":0.96,"supports":true,"quote":null}]},{"id":"b1529814-3b54-429e-b54c-c2d2b5a2bea0","kind":"availability","statement":"The repository contains the Timestral-5M checkpoint.","citations":[{"source":3,"supportProbability":0.96,"supports":true,"quote":{"text":"This repository contains the 5M checkpoint.","rights":"original"}}]},{"id":"cc05232f-63ad-464d-a85b-927e67ed8eb6","kind":"availability","statement":"The 209M checkpoint and source code are publicly available.","citations":[{"source":2,"supportProbability":0.62,"supports":false,"quote":{"text":"We’re releasing the source code in our [GitHub repository](https://github.com/mistralai/timestral/) to make it easy to run Timestral and reproduce our benchmark results.","rights":"original"}}]},{"id":"5b05930d-8f8c-4660-9a57-82d604613c5d","kind":"capability","statement":"The model accepts univariate or multivariate target variates and past-only or known-future covariates.","citations":[{"source":5,"supportProbability":0.93,"supports":true,"quote":{"text":"- **Inputs:** univariate or multivariate target variates, plus past-only and known-future covariates, numeric or categorical (categorical covariates are target-encoded).","rights":"original"}},{"source":6,"supportProbability":0.86,"supports":true,"quote":null}]},{"id":"5038be0c-640f-40a5-b78f-0475116986d5","kind":"availability","statement":"The repository contains the 123M checkpoint.","citations":[{"source":5,"supportProbability":0.99,"supports":true,"quote":{"text":"This repository contains the 123M checkpoint.","rights":"original"}}]},{"id":"98206ad9-f845-4147-a5ab-0c6132b66505","kind":"capability","statement":"The model supports context lengths of up to 32,768 time steps and forecast horizons of up to 1,024 time steps in a single forward pass.","citations":[{"source":5,"supportProbability":0.95,"supports":true,"quote":{"text":"- **Forecast horizon**: Up to 1,024 time steps (32 patches) in a single forward pass.","rights":"original"}}]},{"id":"df61affb-943d-432e-a3d3-1a827b1bccf0","kind":"deployment","statement":"Inference uses PyTorch on CPU or GPU.","citations":[{"source":5,"supportProbability":0.96,"supports":true,"quote":{"text":"- **Inference:** PyTorch on CPU or GPU.","rights":"original"}},{"source":6,"supportProbability":0.9,"supports":true,"quote":null}]},{"id":"9fd9b091-6514-4293-85af-e5fe816a172b","kind":"recommendation-relevant","statement":"The model is released under a Modified MIT license with a restriction against violating third-party rights.","citations":[{"source":5,"supportProbability":0.9,"supports":true,"quote":null},{"source":6,"supportProbability":0.94,"supports":true,"quote":null}]},{"id":"c9ed13ad-9686-4ac1-8dbc-cde97bfa5dee","kind":"capability","statement":"Timestral-123M is a time series foundation model for zero-shot forecasting.","citations":[{"source":5,"supportProbability":0.7,"supports":true,"quote":{"text":"Timestral is a **family of time series foundation models** for zero-shot forecasting, available in [**5M**](https://huggingface.co/mistralai/Timestral-5M), [**28M**](https://huggingface.co/mistralai/Timestral-28M), [**123M**](https://huggingface.co/mistralai/Timestral-123M),…","rights":"original"}}]}]},"nextCursor":null,"license":{"catalog":{"name":"CC BY 4.0","spdx":"CC-BY-4.0","url":"https://creativecommons.org/licenses/by/4.0/","attribution":"wowsuchai","attributionUrl":"https://wowsuchai.withglint.app","covers":["catalog metadata","classifications","generated summaries","feed event metadata"]},"excluded":{"covers":["source quotations","linked source content"],"terms":"Not licensed by wowsuchai. Source quotations and linked source content keep their original rights."}}}