Mercury 3
- Price
- Paid
- API
- Yes
- Weights
- No
WHAT IT DOES
Mercury 3 is Inception's diffusion language model, announced October 19, 2026. Inception says it is more than 40 times larger than Mercury 1, with a 1 million token context, up from 260,000 on Mercury 2.5. The company reports 1,068 tokens per second on Nvidia GPUs and says that is four to eight times faster than speed-optimized frontier models. List price is $0.20 per million input tokens and $0.70 per million output tokens. At launch the price is half of that, and cached input is 90 percent off the input price. It is on the Inception API, OpenRouter, Baseten and Vercel, and speaks the OpenAI API format. Inception says it matches or beats GPT-6 Luna and Claude Haiku 5.5 on a range of benchmarks.
WHY WE PICKED IT
Diffusion language models have mostly been demos. Mercury 3 is one you can call, with a published token price and a 1 million token context. Treat the quality claims as Inception's until an independent index lists it.
QUICK FACTS
STRENGTHS
- ✓1 million token context
- ✓1,068 tokens per second on Nvidia GPUs, per Inception
- ✓$0.20 per million input tokens and $0.70 output, half off at launch
- ✓OpenAI-compatible API, also on OpenRouter, Baseten and Vercel
LIMITATIONS
- ⚠Closed weights
- ⚠Speed and quality comparisons are Inception's, not an independent index
- ⚠The half-price launch rate is temporary
- ⚠A diffusion model is a poor fit if you need the weights on your own hardware
GETTING STARTED
Try the chat demo at chat.inceptionlabs.ai, or create a key at platform.inceptionlabs.ai. Inception includes 100 million free tokens on the API. Point an OpenAI SDK client at Inception's base URL. The launch discount is $0.10 per million input tokens and $0.35 output.
OFFICIAL LINKS
SIMILAR TOOLS
PRICING
Requires a paid subscription.
FEATURED IN GUIDES
Frontier Training Without the Live Web
OpenAI says frontier training will use cached and simulated web content, after a model reached the l...
Open Models: What They Cannot Do
The benchmark gap has largely closed: open models now score within a point of the closed frontier on...
Hardware For Local AI Models: What The Memory Shortage Changed
Capacity decides whether a model runs. Bandwidth decides how fast it talks, and it is the number mos...
How To Run AI Models Locally: Hardware, Quantisation And Engines
Running a frontier open model locally is a memory problem before it is a speed problem, and a bandwi...
Uncensored AI Models: What Abliteration Actually Does To Them
Abliteration removes a model's ability to refuse by deleting one direction from its activations. It ...
EXPLORE ALTERNATIVES
Compare Mercury 3 with 5+ similar llms AI tools.
FREQUENTLY ASKED QUESTIONS
Is Mercury 3 free?
No, Mercury 3 requires a paid subscription.
Does Mercury 3 have an API?
Yes, Mercury 3 offers an API for programmatic integration.
What is Mercury 3 best for?
Mercury 3 is best for Best Diffusion LLM. Mercury 3 is Inception's diffusion language model, announced October 19, 2026. Diffusion language models have mostly been demos. Mercury 3 is one you can call, with a published token price and a 1 million token context. Treat the quality claims as Inception's until an independent index lists it.
What platforms does Mercury 3 support?
Mercury 3 supports web, api.
Is Mercury 3 open source?
No, Mercury 3 is not open source.
How do I get started with Mercury 3?
Try the chat demo at chat.inceptionlabs.ai, or create a key at platform.inceptionlabs.ai. Inception includes 100 million free tokens on the API. Point an OpenAI SDK client at Inception's base URL. The launch discount is $0.10 per million input tokens and $0.35 output.
How do I use Mercury 3?
Mercury 3 is a large language model for text generation, analysis, and conversation. Access through the web interface. Enter prompts or questions to get responses. It excels at 1 million token context.
FEATURED ON CURATEDAI
Work on Mercury 3? You're hand-reviewed in our directory. Add this badge to your site — it links back to this profile.
