Last updated
What is DeepSeek used for in a Bolna voice agent?
DeepSeek is a low-cost LLM option for a Bolna voice agent, and deepseek-v4-flash is the model the Bolna docs recommend for cost-sensitive, high-volume calling. Pick it for simple to moderate conversations such as reminders, lead qualification and COD confirmation; for complex reasoning or strict English instruction following, test it against gpt-5.4-mini.
How do I set up DeepSeek with Bolna?
- Get an API key from DeepSeek and connect it at platform.bolna.ai/auth/deepseek.
- Open your agent in Agent Studio and go to the Intelligence tab.
- Choose DeepSeek as the provider and
deepseek-v4-flashas the model. - Keep the token limit around 150 and the temperature around 0.2 for consistent scripted calls.
The DeepSeek setup guide in the Bolna docs has the API config and migration notes.
Which DeepSeek models does Bolna support?
| Model | Context | Concurrent requests | Best for on Bolna |
|---|---|---|---|
deepseek-v4-flash | 1M tokens | 2,500 | Most voice agents. Recommended: fastest and lowest cost |
deepseek-v4-pro | 1M tokens | 500 | Complex reasoning tasks |
deepseek-chat, deepseek-reasoner | Not applicable | Not applicable | Deprecated July 24, 2026. Migrate to V4 |
What can you build with DeepSeek and Bolna?
High-volume outbound campaigns. The 2,500 concurrent request limit on deepseek-v4-flash fits large batch campaigns. The Bolna docs pair it with Deepgram Nova-2 for cost-sensitive setups; see the Deepgram integration.
Reminders and COD confirmation. Short, predictable calls where DeepSeek's quality is comparable to larger models at a lower price.
Lead qualification. Ask a fixed set of questions across a long contact list at a lower LLM cost per call.
FAQ lines with a long shared prompt. Automatic prompt caching lowers the cost of agents that repeat the same system prompt on every call.