Source: vercel.com/changelog/mistral-large-4-updated-weights-and-lower-pricing-on-ai-gateway
According to Mistral, the updated weights improve agentic coding and defensive cybersecurity, and they use tokens more efficiently on coding tasks. Mistral also reports gains in spreadsheet creation and editing, web research, and resistance to prompt injection.
Developers already calling mistral/mistral-large-4 through AI Gateway need no change, because the existing model ID now maps to the updated weights and pricing. Vercel's changelog includes an AI SDK example using streamText with that ID. Coding agents such as Claude Code, Codex and Cursor can be configured by installing the latest Vercel CLI and running npx vercel ai-gateway setup, then selecting the model.
Vercel says AI Gateway passes provider pricing through with no markup and charges no platform fee on inference, including Bring Your Own Key requests. The gateway also offers retries, failover, custom reporting, Zero Data Retention support and budgets for API keys.
Why it matters if you build agents
Agent builders using AI Gateway can keep the same model string and get lower token costs without a code change.
Teams running Mistral Large 4 for coding or research should recheck their cost estimates, since the changelog does not list exact new prices.
Get stories like this every weekday
One short email with the day's agent tooling news.