View on models.dev Diffusion at Scale: Introducing Mercury 3
Mercury 3 is the most capable diffusion LLM released to date. It outperforms GPT-6 Luna and Claude Haiku 5.5 while running 4–8x faster at 1,000+ tokens per second.
Inception Team
Mercury 3
Inception
A diffusion-based language model for reasoning, tool use, and latency-sensitive agentic workflows, served through Inception's API.
Specs
Model
- Family
- Mercury
- Variant
- 3
- Weights
- Proprietary
Limits
- Context
- 1,000,000 tokens
- Max output
- 131,072 tokens
Modalities
- Input
- Text
- Output
- Text
Pricing
- Input
- $0.1 / 1M
- Output
- $0.35 / 1M
- Cache read
- $0.01 / 1M
Capabilities
- Reasoning
- Yes
- Reasoning control
- Effort levels
- Tool calling
- Yes
- Attachments
- No
Dates
- Released
- 16 Oct 2026
- Updated
- 20 Oct 2026
Articles
All articlesinceptionlabs.ai