Pixtral Large 2411
Pixtral Large combines a 123B text decoder with a 1B vision encoder. Built on Mistral Large 2, it reads documents, charts and natural images within a 128K context.
Deprecated API; check access details
A separate multimodal checkpoint
The vision component makes this a different trained model from the text-only Large 24.11 released alongside it. The launch describes at least 30 high-resolution images fitting within its context under the stated setup. Image resolution and text still share that context budget.Mistral Large 2.1 primary source
Research download and commercial terms
Pixtral-Large-Instruct-2411 is distributed under the Mistral Research License. Its card limits that license to research purposes and directs other uses to Mistral for separate permission. The public availability of weights does not make them Apache-licensed.Pixtral Large 2411 official model card
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Deprecated API; check access details |
| Launch date | |
| Access | API, Downloadable |
| Model identifiers | Pixtral-Large-Instruct-2411pixtral-large-2411 |
| Context window | 128,000 tokens The model catalog advertises 128k context for this version. This is not a measured recall limit. Pixtral Large 2411 model and lifecycle record |
| Weight license | Mistral Research License; commercial use needs separate permission/licensePixtral Large 2411 official model card |
| Access and lifecycle | Hosted API deprecated on 27 February 2026. Downloadable weights remain listed.Pixtral Large 2411 model and lifecycle recordPixtral Large 2411 official model card |
| Parameters | 123B decoder plus 1B vision encoderMistral Large 2.1 primary source |