Mercury 3
Mercury 3 is Inception's diffusion language model, which writes many tokens in parallel rather than one at a time, reaching over 1,000 tokens per second on NVIDIA GPUs with a 1M-token context, tool calling and structured outputs. It has an OpenAI-compatible API and is also offered on OpenRouter, Baseten and Vercel, which makes it a fit for voice bots, support agents and browser agents where response speed matters. Pricing is per million input and output tokens.
Updated
✨ Try AgentScout's free AI image generator →Good for: browsing the web, building voice agents, customer support
Official channels
Use cases
About Mercury 3
Mercury 3 is listed in our models & local runtimes category and is aimed at builders. Mercury 3 is Inception's diffusion language model, which writes many tokens in parallel rather than one at a time, reaching over 1,000 tokens per second on NVIDIA GPUs with a 1M-token context, tool calling and structured outputs. It has an OpenAI-compatible API and is also offered on OpenRouter, Baseten and Vercel, which makes it a fit for voice bots, support agents and browser agents where response speed matters. Pricing is per million input and output tokens. Pricing, as the maker states it: $0.20 / $0.70 per million input / output tokens. Maturity: Active. We last checked the official link on 2026-10-22.
Mercury 3 at a glance
- Category: Models & local runtimes
- Built for: Developers and teams building with AI agents
- Pricing: $0.20 / $0.70 per million input / output tokens (as the maker states it)
- Official site: inceptionlabs.ai
- Listed on AgentScout: Oct 22, 2026
Mercury 3 FAQ
What is Mercury 3?
Mercury 3 is Inception's diffusion language model, which writes many tokens in parallel rather than one at a time, reaching over 1,000 tokens per second on NVIDIA GPUs with a 1M-token context, tool calling and structured outputs. It has an OpenAI-compatible API and is also offered on OpenRouter, Baseten and Vercel, which makes it a fit for voice bots, support agents and browser agents where response speed matters. Pricing is per million input and output tokens.
How much does Mercury 3 cost?
Mercury 3's pricing, as the maker states it: $0.20 / $0.70 per million input / output tokens. Check the official site for current plans.
What are alternatives to Mercury 3?
Similar tools include LocalAI, Oumi, Meta Muse Wonder, GLM-5, pooled and MiroThinker.
Facts from our link-checked listing and the official site. See something wrong? Suggest an edit.
Mercury 3 ratings and reviews
Be the first to rate Mercury 3: tap 1 to 5 stars below. A written review is optional.
Mercury 3 alternatives
Other models & local runtimes tools in our directory. See all models & local runtimes tools →
LocalAI
Self-hosted engine with an OpenAI-compatible API for running LLM, vision, voice and image models on your own hardware, including agents.
Oumi
Open platform for fine-tuning, evaluating and deploying open agentic LLMs and VLMs.
Meta Muse Wonder
Meta's new models for personal and professional agent tasks, available via Meta Model API and Muse Code.
GLM-5
Z.ai's open-weights GLM-5 model family built for coding and long-horizon agentic tasks, also offered through the Z.ai API.
pooled
Peer-to-peer LLM inference in the browser: pool your laptops, phones and other devices into a room to run large open models for chat and coding agents, with no GPU server.
MiroThinker
Deep research agent and open models tuned for complex research and forecasting tasks.
Popular searches with this tool
🏆 Ranked in: LocalAI alternatives
🔁 Listed in: LocalAI alternatives
🧭 Task guides: Voice and phone agents for customer support · AI browser automation tools
💲 Estimate API cost: compare per-request and monthly costs for GPT, Claude, Gemini and Grok models.
Badge for makers
Built Mercury 3? Add this badge to your README or site. It links back to this page. About badges
