Devstral 2 2512

Release Date: Dec 9, 2025     Creator: Mistral AI

This model has been removed from VividLLM and is no longer available for use.

Devstral 2512 is a powerful code agent model built by Mistral AI for complex software engineering tasks.

Access Devstral 2 2512 securely with complete privacy. Our local-first approach means your text chats and prompts are stored exclusively on your browser via IndexedDB, never on a remote server. When you delete a chat, it is permanently wiped locally, and we ensure zero prompt training is used across all model endpoints.

Try Devstral 2 2512 instantly

Here

🔒 Requires a secure session. Sign in with Google to protect your data.

Model Specifications
Knowledge Cutoff
early 2025
N/A
Context (VividLLM)
32,000 Tokens
N/A
With a context window of 32,000 tokens, this model can 'remember' and analyze up to 48 pages of data at once. This makes it a top choice for reviewing legal contracts and dense academic chapters.
Context (Native)
262,144 Tokens
N/A
Input ModalitiesTextImagePDFTextImagePDF
Tool Calling
yes
N/A
Output TypeTextText
Default Max Output81928192
Input Weight
2.000 x (Casual)
N/A
Output Weight
2.000 x (Casual)
N/A

Disclaimer: We are providing Image Upload options to only limit models as of now, so some models which have image upload options might now have that feature on our site.

Devstral 2 2512: Secure & Private Access | No Prompt training