# Baseten adds deferred tool loading to open models, cutting agent costs by up to 62%

> The new feature reduces token usage and speeds up tasks, saving up to 62.5% in cost and 26% in time for coding agents.

Oossa · 2026-11-19 · https://oossa.com/en/baseten-adds-deferred-tool-loading-to-open-models-cutting-agent-costs-by-up-to-6

Baseten announced that its Model APIs now support deferred tool loading for open models such as DeepSeek, Kimi and GLM. The change keeps tool schemas out of the system prompt and loads them only when needed. In Claude Code tests with three registered plugins, coding benchmarks became 62.5% cheaper and 26.2% faster. Agentic benchmarks also saw savings of about 51% in cost and 15% in time.

## The facts

- Baseline cost reduction: 62.5% cheaper on coding benchmarks (Baseten).
- Speed improvement: 26.2% faster on coding benchmarks (Basenen).
- Announcement date: November 19, 2026.

## Why it matters

Developers can run AI agents cheaper and faster without changing their models or code.

## Sources & references

1. [Deferred tool loading: Make your agent 62.5% cheaper without changing your model or harness](https://www.baseten.co/blog/deferred-tool-loading-without-changing-model-harness/) – Baseten

Last updated: 2026-11-19
