volodia

We're not publicly announcing the model size right now, but we can compare against other models with a similar intelligence -- what we see is a big boost in speed for the same level of capabilities.

We are going to be sharing additional information like task completion times in the near future as the model undergoes additional benchmarking by third parties. We also plan to share data on specific use cases where the model traces new points of the Pareto frontier and and speed cost.

For a preview of the latter, check out our analysis of the Mercury Voice model: https://www.inceptionlabs.ai/blog/introducing-mercury-voice

Not at the moment, but we are looking to share more information about the model in the future via technical blog posts and technical reports.

This is an area that we're really excited about. We have internal prototypes, and if you're interested in trying it out as an early tester, please reach out to hello@inceptionlabs.ai

We have internal prototypes, so please reach out to hello@inceptionlabs.ai if you're interested in early access

We might not be able to precisely explain the 6% difference on that specific benchmark, but overall we have found the model to be Haiku- or Luna-level in our own daily usage. That means that it won't yet replace Opus or Fable as the primary driver of a harness, but we have used it successfully for subagents, including implementing kernel optimization ideas produced by a larger model (and those improvements have gone into production -- a form of RSI).

In general, the model continues to improve when given additional high quality data like a normal LLM, which is what makes us very bullish on the scaling of dLLMs.

We've been happy with how performance has scaled with batch size. The model runs in our API at reasonable costs, which is also what allows us to serve the model with competitive prices ($0.10 per 1M input and $0.35 per 1M output at launch).

We're not releasing the weights, but the model is available using a standard API, and it integrates with your favorite harnesses.

Here is a page that compiles the list of integrations: https://docs.inceptionlabs.ai/integrations/overview

For example, it's just a few small steps to run the model in Cursor: https://docs.inceptionlabs.ai/integrations/cursor

You shouldn't expect Opus-level performance yet, but early testing has found the model to be great for coding subagents.

Hi HN, this is Volodymyr, Inception Co-Founder & Chief Scientist.

Today we're launching Mercury 3, our most capable production diffusion LLM.

Quality: Outperforms GPT-6 Luna and Claude Haiku 5.5 across a range of benchmarks at matched reasoning effort. Speed: 1,000+ tokens per second on widely available NVIDIA Blackwell GPUs, 4–8x faster than those models. Context: 1M tokens, up from 260K. Price: $0.10 per million input and $0.35 per million output tokens at launch.

The main takeaway is that diffusion scales. Since Mercury 1 we’ve scaled the model more than 40x in total parameters, and intelligence has gone up at every step without giving up speed. Diffusion generates tokens in parallel through iterative denoising rather than one at a time, so as reasoning models think longer per answer, the speed advantage over autoregressive models of the same size grows.

Early-access partners are running Mercury 3 in browser agents, tutoring apps, real-time search, and customer support. It’s OpenAI API compatible, and it’s also available on OpenRouter, Baseten, and Vercel.

Try it on our API with 100M free tokens.

Happy to dig into the architecture, scaling results, evals, or anything else in the comments.

Thank you for the detailed feedback! I shared this already with the team.

This looks like an inference glitch that we are working on fixing, thank you for flagging.

There are many ways to do it, but the simplest approach is block diffusion: https://m-arriola.com/bd3lms/

There are also more advanced approaches, for example FlexMDM, which essentially predicts length of the "canvas" as it "paints tokens" on it.

Would love to hear about your experience. Send us an email.

Not imminently, but hard to predict where the field will go

There are few: fast agents, deep research, real-time voice, coding. The other thing is that when you have a fast reasoning model, you spend more effort on thinking in the same latency budget, which pushed up quality.

We agree! In fact, there is an emerging class of models aimed at fast agentic iteration (think of Composer, the Flash versions of proprietary and open models). We position Mercury 2 as a strong model in this category.

That is also our view! We see Mercury 2 as enabling very fast iteration for agentic tasks. A single shot at a problem might be less accurate, but because the model has a shorter execution time, it enables users to iterate much more quickly.

You can think of Mercury 2 as roughly in the same intelligence tier as other speed-optimized models (e.g., Haiku 4.5, Grok Fast, GPT-Mini–class systems). The main differentiator is latency — it’s ~5× faster at comparable quality.

We’re not positioning it as competing with the largest models (Opus 4.5, etc.) on hardest-case reasoning. It’s more of a “fast agent” model (like Composer in Cursor, or Haiku 4.5 in some IDEs): strong on common coding and tool-use tasks, and providing very quick iteration loops.

Thanks for trying it and for the thoughtful feedback, really appreciate it. And we’re actively working on improving quality further as we scale the models.

Thank you for your patience. We are working to handle the surge in demand.

Just to clarify one point: Mercury (the original v1, non-reasoning model) is already used in production in mainstream IDEs like Zed: https://zed.dev/blog/edit-prediction-providers

Mercury v1 focused on autocomplete and next-edit prediction. Mercury 2 extends that into reasoning and agent-style workflows, and we have editor integrations available (docs linked from the blog). I’d encourage folks to try the models!

I’d push back a bit on the Pareto point.

On speed/quality, diffusion has actually moved the frontier. At comparable quality levels, Mercury is >5× faster than similar AR models (including the ones referenced on the AA page). So for a fixed quality target, you can get meaningfully higher throughput.

That said, I agree diffusion models today don’t yet match the very largest AR systems (Opus, Gemini Pro, etc.) on absolute intelligence. That’s not surprising: we’re starting from smaller models and gradually scaling up. The roadmap is to scale intelligence while preserving the large inference-time advantage.

Co-founder / Chief Scientist at Inception here. If helpful, I’m happy to answer technical questions about Mercury 2 or diffusion LMs more broadly.

There is also this one that was released in October: https://github.com/kuleshov/char-mdlm

the LLaDA paper is a scaled-up version of this paper; they cite it as an anonymous ICLR submission

Great question! The model can more efficiently leverage existing GPU hardware---it performs more computation per unit of memory transferred; this means that on older hardware one should be able to get similar inference speeds as one would get on recent hardware with a classical LLM. This is actually interesting commercially, since it opens new ways of reducing AI inference costs.

Yes, we plan to be releasing a tech report soon. We are not open sourcing the models at launch time, but we have a roadmap of future releases in which we hope to make some of our models accessible to the research community.

That's a good point. In this context, we've been using "commodity GPUs" to refer to standard Nvidia hardware, in contrast to specialized chips like Groq and Cerebras. While these chips also achieve fast speeds, they are not nearly as ubiquitous as Nvidia GPUs. We think that matching their performance on standard Nvidia hardware can make AI much more affordable. We also support any GPUs, not just H100's.

We're going to be releasing a tech report soon, stay tuned!

Good question! We are not open sourcing the models at launch time, but we have a roadmap of future releases in which we hope to make some of our models accessible to the research community.

The short answer is that we do more than one parallel pass over multiple tokens: we iteratively refine them over a few passes to fix incoherences. This can be seen as a generalization of diffusion algorithms that underlie systems like Midjourney or Sora.