Mercury Models by Inception Labs
1 model2026Up to 131k ctxFrom $0.25/1M input
About
Inception Labs' Mercury series of diffusion-based large language models (dLLMs). Mercury uses a fundamentally different architecture from autoregressive models, enabling faster generation speeds. Mercury 2 is a commercial-scale reasoning model optimized for code and analysis tasks.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
1 in view
Mercury 2Current
Use when the workload needs 131k context and structured outputs.
2026-02131k contextstructured outputs
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Mercury 2 | Use when the workload needs 131k context and structured outputs. | 2026-02 | 131k contextstructured outputs | Current |
Release Timeline
1 release group2026-02
1 current
Mercury 2
Current131k contextstructured outputs
Specifications(1 models)
| Model | Released | Context | Structured Outputs |
|---|---|---|---|
| Mercury 2 | 2026-02 | 131k | Yes |
Available From(2 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Mercury 2 | OpenRouter | $0.25 | $0.75 | Serverless |
| Mercury 2 | Vercel AI Gateway | $0.25 | $0.75 | Serverless |
Frequently Asked Questions
- What is Mercury used for?
- Mercury is used for structured outputs and coding. The family description and listed model capabilities point to those workloads as the best fit.
- How does Mercury compare to Claude 3?
- Mercury by Inception Labs is strongest where you need structured outputs, while Claude 3 by Anthropic is the closest related family to check for vision and multimodal work. Mercury has 1 listed variant and reaches up to 131k context, while Claude 3 reaches up to 200k context, so compare the specs and pricing tables before choosing a production model.