Inception's diffusion reasoning model for chat, multi-step problem solving, and tool use
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...
Context
1M
1,000,000 tokens
Max output
131K
131,072 tokens
Input
$0.10
per 1M tokens
Output
$0.35
per 1M tokens
Providers
5
5 listings
Intelligence
—
Not yet scored
Pricing
USD per 1M tokens · listed by Inception
| Token type | Standard |
|---|---|
| Input | $0.10 |
| Output | $0.35 |
| Cache read | $0.01 |
Lowest input price across providers: $0.10.
Capabilities
ReasoningYes
Effort: low, medium, high
Tool callYes
Structured outputYes
TemperatureYes
AttachmentsNo
Interleaved reasoningNo
Open weightsNo
ModeratedNo
Input
Text
Output
Text
Benchmarks
Published scores and Design Arena Elo
- IntelligenceNot scored—
- CodingNot scored—
- AgenticNot scored—
- Arena1,215#35 of 128
Rankings
Position on each OneAtlas leaderboard
Providers · 5
5 listings
| Provider | Model ID | Context | Input limit | Output | Input $ | Output $ | Reasoning $ | Cache read | Cache write | Audio in | Audio out | Reasoning | Tools | Structured | Temp. | Updated |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Inception | mercury-3 | 1M | — | 131K | $0.10 | $0.35 | — | $0.01 | — | — | — | Yes | Yes | Yes | Yes | Oct 20, 2026 |
 Kilo Gateway | inception/mercury-3 | 1M | — | 131K | $0.20 | $0.70 | — | $0.02 | — | — | — | Yes | Yes | Yes | Yes | Oct 16, 2026 |
 NanoGPT | inception/mercury-3 | 1M | 1M | 131K | $0.10 | $0.35 | — | $0.01 | — | — | — | Yes | Yes | Yes | Yes | Oct 20, 2026 |
 Venice AI | mercury-3 | 1M | — | 131K | $0.10 | $0.35 | — | $0.01 | — | — | — | Yes | Yes | Yes | Yes | Oct 20, 2026 |
 Vercel AI Gateway | inception/mercury-3 | 1M | — | 131K | $0.10 | $0.35 | — | $0.01 | — | — | — | Yes | Yes | Unknown | Unknown | Oct 19, 2026 |