Inception: Mercury 3
by InceptionInception: Mercury 3 is a large language model from Inception. It costs $0.100 per million input tokens and $0.350 per million output tokens. Its context window is 1M tokens.
- Input / 1M tokens
- $0.100
- Output / 1M tokens
- $0.350
- Cached input / 1M
- $0.010
- Context window
- 1M
On repeated prefixes
tokens
Who serves it cheapest
1 hosts serve Mercury 3. Same weights, same API - the price difference is pure margin and routing.
| Provider | Input / 1M | Output / 1M | Context | Throughput | Uptime 24h |
|---|---|---|---|---|---|
| InceptionCheapest | $0.100 | $0.350 | 1M | - | 99.7% |
About Mercury 3
Mercury 3 is a diffusion-based language model from Inception, and the lab's most capable model to date. It is suited for agentic workflows and real-time applications where latency compounds, such...
Specifications
| Model ID | inception/mercury-3 |
|---|---|
| Provider | Inception |
| Context window | 1M tokens |
| Max output | 131K tokens |
| Input modalities | text |
| Output modalities | text |
| Knowledge cutoff | - |
| Open weights | No |
| Released | October 16, 2026 |
Cheaper alternatives
Models that cost less than Mercury 3 while keeping at least half its context window and every input modality it supports.
Frequently asked
How much does Inception: Mercury 3 cost?
$0.100 per million input tokens and $0.350 per million output tokens. Cached input reads cost $0.010 per million tokens.
What is the context window of Inception: Mercury 3?
1M tokens, with up to 131K tokens of output per request.