# Mercury 3

> Mercury 3 is a proprietary language model from Inception Labs, released on 16 Oct 2026. It accepts text and generates text, with a 1M-token context window and up to 131K output tokens per response. As of 21 Oct 2026, the lowest listed API price is $0.10 per million input tokens and $0.35 per million output tokens (Inception API), across 3 providers we track.

Source page: https://www.aimodel.directory/models/mercury-3
Last verified: 21 Oct 2026

## Specifications

- **Developer:** Inception Labs
- **Release date:** 16 Oct 2026 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Weights:** Proprietary (API only) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Context window:** 1M (1,000,000 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Max output:** 131K tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Input:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Output:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Reasoning mode:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Tool calling:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)
- **Structured output:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/inception/models/mercury-3.toml)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|
| Inception API | $0.10 | $0.35 | $0.01 | 1M |
| OpenRouter | $0.10 | $0.35 | $0.01 | 1M |
| Vercel AI Gateway | $0.10 | $0.35 | $0.01 | 1M |

## FAQ

### What is the context window of Mercury 3?

Mercury 3 has a context window of 1,000,000 tokens (1M) and can generate up to 131,072 tokens in a single response, according to models.dev.

### How much does the Mercury 3 API cost?

Through Inception API, Mercury 3 costs $0.10 per million input tokens and $0.35 per million output tokens. Prices last verified 21 Oct 2026.

### Is Mercury 3 open source?

No. Mercury 3 is a proprietary model; Inception Labs has not released its weights. It is available through APIs including Inception API, OpenRouter, Vercel AI Gateway.

### When was Mercury 3 released?

Inception Labs released Mercury 3 on 16 Oct 2026, according to models.dev.

### Which providers offer Mercury 3?

We track Mercury 3 on 3 providers: Inception API, OpenRouter, Vercel AI Gateway.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "Mercury 3", https://www.aimodel.directory/models/mercury-3