Run a capable Microsoft small language model locally or inside your own AI application.
Text & Writing
All tools
View more
Free
Free tier
Access
Listed as Inception Labs currently presents Mercury 2 and Mercury Edit 2 as diffusion language models with API pricing of $0.25 input and $0.75 output per one million tokens, plus an early-access route and a temporary free-token promotion. No permanent no-cost plan was verified. It is an API and model platform rather than an unlimited free chatbot.
0
Coding & Dev Checked Aug 6, 2026
Mercury 2
Use a fast diffusion language model when latency matters in a coding or AI application.
Best for: Developers evaluating high-throughput text, coding, agent, or voice workflows through an API.
What you can do with it
- Prototype a fast coding assistant
- Test low-latency agent responses
- Compare token speed and cost with other APIs
- Build a voice workflow that needs quick text generation
- Try Mercury Edit for code changes
Good to know
- Current Mercury 2 family
- Parallel diffusion generation approach
- API-compatible integration
- Targets coding and voice latency
Limitations
- No permanent free plan verified
- Early-access availability can change
- API billing still applies after any promotion
Other official links
Keep exploring
Similar tools
FreeAn open-source mixture-of-experts language model with 80B total parameters and 13B active parameters for reasoning, chat, coding, and text generation.
AI Agents & Chatbots
Free tierA Mistral AI workspace that combines chat, search, creation, and an agentic coding mode in one place.
Coding & Dev