I
Mercury 3
Mercury 3, a language model from Inception.
modelOverride: { model: "mercury-3" }
Inception·Released October 2026Pricing
- Input
- $0.100 / 1M
- Output
- $0.35 / 1M
- Cached input
- $0.010 / 1M
Specifications
- Context window
- 1,000,000 tokens
- Max output
- 131,072 tokens
- Released
- October 2026
- Capabilities
- FastLarge contextReasoning
Parameters
Model options, passed in
modelOverride.config.| Parameter | Type | Default | Range / options |
|---|---|---|---|
reasoning_effortHow long the model thinks before it answers. Instant is fastest; High is for complex, multi-step problems. | Select | low | instantlowmediumhigh |
Use this model
await mindstudio.generateText({ message: "...", modelOverride: { model: "mercury-3", config: { reasoning_effort: "low", }, }, });