Inception's Mercury 3 Now Available on AI Gateway

What happened

Inception's Mercury 3 has been made available on AI Gateway, offering users a 50% discount during its launch phase. The model is now accessible through the platform's centralized API, allowing for seamless integration and usage.

Mercury 3 is designed to provide enhanced reasoning capabilities through diffusion techniques, enabling parallel refinement of text generation. Users can test the model in the AI Gateway model playground before deploying it in production environments.

Why it matters

The availability of Mercury 3 on AI Gateway expands access to high-performance language models, particularly for developers and enterprises looking for scalable and cost-effective solutions. The 50% discount during the launch phase may encourage early adoption and experimentation with the model.

Key details

Mercury 3 achieves generation speeds exceeding 1,000 tokens per second and supports a context window of 1 million tokens. These features make it suitable for handling extended documents and lengthy conversation histories.

To use Mercury 3, users must set the AI_GATEWAY_API_KEY in their environment and specify the model name as inception/mercury-3. AI Gateway also provides tools for monitoring usage, managing costs, and configuring performance settings.