7 ms·
Mercury 2: Fast reasoning LLM powered by diffusion
- dvt 8mo agoWhat excites me most about these new 4figure/second token models is that you can essentially do multi-shot prompting (+ nudging) and the user doesn't even feel it, potentially fixing some of the weird hallucinatory/non-deterministic behavior we sometimes end up with.
- volodia 8mo agoThat 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.
- lostmsu 8mo agoRegular models are very fast if you do batch inference. GPT-OSS 20B gets close to 2k tok/s on a single 3090 at bs=64 (might be misremembering details here).
- rahimnathwani 8mo agoRight but everyone else is talking about latency, not throughput.
- tl2do 8mo agoGenuine question: what kinds of workloads benefit most from this speed? In my coding use, I still hit limitations even with stronger models, so I'm interested in where a much faster model changes the outcome rather than just reducing latency.
- irthomasthomas 8mo agomulti-model arbitration, synthesis, parallel reasoning etc. Judging large models with small models is quite effective.
- layoric 8mo agoI think it would assist in exploiting exploring multiple solution spaces in parallel, and can see with the right user in the loop + tools like compilers, static analysis, tests, etc wrapped harness, be able to iterate very quickly on multiple solutions. An example might be, "I need to optimize this SQL query" pointed to a locally running postgres. Multiple changes could be tested, combined, and explain plan to validate performance vs a test for correct results. Then only valid solutions could be presented to developer for review. I don't personally care about the models 'opinion' or recommendations, using them for architectural choices IMO is a flawed use as a coding tool. It doesn't change the fact that the most important thing is verification/validation of their output either from tools, developer reviewing/making decisions. But even if don't want that approach, diffusion models are just a lot more efficient it seems. I'm interested to see if they are just a better match common developer tasks to assist with validation/verification systems, not just writing (likely wrong) code faster.
- cjbarber 8mo agoI've tried a few computer use and browser use tools and they feel relatively tok/s bottlenecked. And in some sense, all of my claude code usage feels tok/s bottlenecked. There's never really a time where I'm glad to wait for the tokens, I'd always prefer faster.
- quotemstr 8mo agoOnce you make a model fast and small enough, it starts to become practical to use LLMs for things as mundane as spell checking, touchscreen-keyboard tap disambiguation, and database query planning. If the fast, small model is multimodal, use it in a microwave to make a better DWIM auto-cook. Hell, want to do syntax highlighting? Just throw buffer text into an ultra-fast LLM. It's easy to overlook how many small day-to-day heuristic schemes can be replaced with AI. It's almost embarrassing to think about all the totally mundane uses to which we can put fast, modest intelligence.
- volodia 8mo agoThere 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.
- corysama 8mo agoCoding auto-complete?
- storus 8mo agoI'd say using them as draft models for some strong AR model, speeding it up 3x. Diffusion generates a bunch of tokens extremely fast, those can be then passed over to an AR model to accept/reject instead of generating them.
- cjbarber 8mo agoIt could be interesting to do the metric of intelligence per second. ie intelligence per token, and then tokens per second My current feel is that if Sonnet 4.6 was 5x faster than Opus 4.6, I'd be primarily using Sonnet 4.6. But that wasn't true for me with prior model generations, in those generations the Sonnet class models didn't feel good enough compared to the Opus class models. And it might shift again when I'm doing things that feel more intelligence bottlenecked. But fast responses have an advantage of their own, they give you faster iteration. Kind of like how I used to like OpenAI Deep Research, but then switched to o3-thinking with web search enabled after that came out because it was 80% of the thoroughness with 20% of the time, which tended to be better overall.
- nubg 8mo agoInteresting perspective. Perhaps also the user would adopt his queries knowing he can only to small (but very fast) steps. I wonder who would win!
- josephg 8mo agoYeah I agree with this. We might be able to benchmark it soon (if we can’t already) but asking different agentic code models to produce some relatively simple pieces of software. Fast models can iterate faster. Big models will write better code on the first attempt, and need less loop debugging. Who will win? At the moment I’m loving opus 4.6 but I have no idea if its extra intelligence makes it worth using over sonnet. Some data would be great!
- estsauver 8mo agoFor what it's worth, most people already are doing this! Some of the subagents in Claude Code (Explore, I think even compaction) default to Haiku and then you have to manually overwrite it with an env variable if you want to change it. Imagine the quality of life upgrade of getting compaction down to a few second blip, or the "Explore" going 20 times faster! As these models get better, it will be super exciting!
- embedding-shape 8mo ago
- ilaksh 8mo agoIt seems like the chat demo is really suffering from the effect of everything going into a queue. You can't actually tell that it is fast at all. The latency is not good. Assuming that's what is causing this. They might show some kind of feedback when it actually makes it out of the queue.
- volodia 8mo agoThank you for your patience. We are working to handle the surge in demand.
- mhitza 8mo agoComment retracted. My bad, missed some details.
- selcuka 8mo agoI think your comment is a bit unfair. > no reasoning comparison Benchmarks against reasoning models: https://www.inceptionlabs.ai/blog/introducing-mercury-2 https://www.inceptionlabs.ai/blog/introducing-mercury-2 > no demo https://chat.inceptionlabs.ai/ https://chat.inceptionlabs.ai/ > no info on numbers of parameters for the model This is a closed model. Do other providers publish the number of parameters for their models? > testimonials that don't actually read like something used in production Fair point.
- mhitza 8mo agoYou are right edited my post (twice actually). Missed the chat first time around (though its hard to see it as a reasoning model when chain of thought is hidden, or not obvious. I guess this is the new normal), and also missed the reasoning table because text is pretty small on mobile and I thought its another speed benchmark.
- selcuka 8mo agoI tried their chat demo again, and if you set reasoning effort to "High", you sometimes see the chain of thought before the answer (click the "Thought for n seconds" text to expand it). That being said, the chain is pretty basic. It's possible that they don't disclose the full follow-up prompt list.
- volodia 8mo agoJust 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 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!
- nylonstrung 8mo agoI'm not sold on diffusion models. Other labs like Google have them but they have simply trailed the Pareto frontier for the vast majority of use cases Here's more detail on how price/performance stacks up https://artificialanalysis.ai/models/mercury-2 https://artificialanalysis.ai/models/mercury-2
- volodia 8mo agoI’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.
- ainch 8mo agoThis understates the possible headroom as technical challenges are addressed - text diffusion is significantly less developed than autoregression with transformers, and Inception are breaking new ground.
- nylonstrung 8mo agoVery good point- if as much energy/money that's gone into ChatGPT style transformer LLMs were put into diffusion there's a good chance it would outperform in every dimension
- nylonstrung 8mo agoI changed my mind: this would be perfect for a fast edit model ala Morph Fast Apply https://www.morphllm.com/products/fastapply https://www.morphllm.com/products/fastapply It looks like they are offering this in the form of "Mercury Edit"and I'm keen to try it
- arjie 8mo agoPlease pre-render your website on the server. Client-side JS means that my agent cannot read the press-release and that reduces the chance I am going to read it myself. Also, day one OpenRouter increases the chance that someone will try it.
- dhruv3006 8mo agoI am little underwhelmed by anything diffusion at the moment - they didn't really deliver.
- volodia 8mo agoCo-founder / Chief Scientist at Inception here. If helpful, I’m happy to answer technical questions about Mercury 2 or diffusion LMs more broadly.
- CamperBob2 8mo agoSeems to work pretty well, and it's especially interesting to see answers pop up so quickly! It is easily fooled by the usual trick questions about car washes and such, but seems on par with the better open models when I ask it math/engineering questions, and is obviously much faster.
- volodia 8mo agoThanks 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.
- kristianp 8mo agoHow big is Mercury 2? How many tokens is it trained on? Is it's agentic accuracy good enough to operate, say, coding agents without needing a larger model to do more difficult tasks?
- volodia 8mo agoYou 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.
- nayroclade 8mo agoIs the approach fundamentally limited to smaller models? Or could you theoretically train a model as powerful as the largest models, but much faster?
- MarcLore 8mo ago[dead]
- lprimeisafk 8mo agoIt fails the car wash test
- emi2k01 8mo agoThe test: > The car wash is 100m away from my house. Should I walk or drive? I didn't find a way to share the conversation, so here it is the full response in `medium` settings: --- Walking 100 m is a very short distance—roughly a minute or two at a normal pace. In most situations, walking will be quicker overall because you avoid the time needed to get into the car, start it, and find parking. It also saves fuel and is better for the environment. *Consider these factors before deciding:* | Factor | Walking | Driving | |--------|---------|---------| | *Time* | ~1–2 min (plus a few seconds to step out) | ~1 min driving, plus ~1–2 min to get in/out and park | | *Energy/Fuel* | None | Uses a small amount of fuel | | *Weather* | Comfortable if dry & mild | Better if it’s raining, very hot, or icy | | *Physical ability* | Easy for most people | Not needed for such a short trip | | *Car availability* | Not required | Must have a car nearby and be willing to use it | | *Environmental impact* | Zero emissions | Small emissions | *Recommendation:* If the weather is pleasant and you don’t need the car for anything else right away, walking is the simplest, fastest, and most eco‑friendly choice. Drive only if you’re dealing with inclement weather, have heavy items to carry, or need the car immediately after the wash. Do you have any specific constraints (e.g., rain, heavy bags, time pressure) that might affect the decision?
- rtfeldman 8mo agoIf a stranger asks me, "Should I walk or drive to this car wash?" then I assume they're asking in good faith and both options are reasonable for their situation. So it's a safe assumption that they're not going there to get their car washed. Maybe they're starting work there tomorrow, for example, and don't know how pedestrian-friendly the route is. Is the goal behind evaluating models this way to incentivize training them to assume we're bad-faith tricksters even when asking benign questions like how best to traverse a particular 100m? I can't imagine why it would be desirable to optimize for that outcome. (I'm not saying that's your goal personally - I mean the goal behind the test itself, which I'd heard of before this thread. Seems like a bad test.)
- chriskanan 8mo agoI can see some promise with diffusion LLMs, but getting them comparable to the frontier is going to require a ton of work and these closed source solutions probably won't really invigorate the field to find breakthroughs. It is too bad that they are following the path of OpenAI with closed models without details as far as I can tell.
- dw5ight 8mo agothis looks awesome!!
- nowittyusername 8mo agoNice, I'm excited to try this for my voice agent, at worst it could be used to power the human facing agent for latency reduction.
- volodia 8mo agoWould love to hear about your experience. Send us an email.
- exabrial 8mo agoI believe Jimmy Chat is still faster by an order of magnitude…
- poly2it 8mo agoWhat does Jimmy Chat have to do with diffusion models?
- serjester 8mo agoThere's a potentially amazing use case here around parsing PDFs to markdown. It seems like a task with insane volume requirements, low budget, and the kind of thing that doesn't benefit much from autoregression. Would be very curious if your team has explored this.
- davistreybig 8mo agoThis is unbelievably fast
- rancar2 8mo agoMy attempt with trying one of their OOTB prompts in the demo https://chat.inceptionlabs.ai https://chat.inceptionlabs.ai resulted in: "The server is currently overloaded. Please try again in a moment." And a pop-up error of: "The string did not match the expected pattern." That happened three times, then the interface stopped working. I was hoping to see how this stacked up against Taalas demo, which worked well and was so fast every time I've hit it this past week.
- dmix 8mo agoI tried Mercury 1 in Zed for inline completions and it was significantly slower than Cursors autocomplete. Big reason why I switched backed to Cursor(free)+Claude Code
- vicchenai 8mo agoThe iteration speed advantage is real but context-specific. For agentic workloads where you're running loops over structured data -- say, validating outputs or exploring a dataset across many small calls -- the latency difference between a 50 tok/s model and a 1000+ tok/s one compounds fast. What would take 10 minutes wall-clock becomes under a minute, which changes how you prototype. The open question for me is whether the quality ceiling is high enough for cases where the bottleneck is actually reasoning, not iteration speed. volodia's framing of it as a "fast agent" model (comparable tier to Haiku 4.5) is honest -- for the tasks that fit that tier, the 5x speed advantage is genuinely interesting.
- smusamashah 8mo agoDoes it mean if it was embedded on a Talaas chip, it could generate ~50,000+ tokens per second?
- Havoc 8mo agoThink pretty much anything is going to get a enormous speed boost if the model isn’t undergoing mem latency but is just inherently baked into the circuits asic style
- Ross00781 8mo agoDiffusion-based reasoning is fascinating - curious how it handles sequential dependencies vs traditional autoregressive. For complex planning tasks where step N heavily depends on steps 1-N, does the parallel generation sometimes struggle with consistency? Or does the model learn to encode those dependencies in a way that works well during parallel sampling?
- herlon214 8mo agoThis looks really nice. When will it be available on OpenRouter?
- alflex 8mo ago[dead]
- alflex 8mo ago[dead]
- nivcmo 8mo ago[dead]
- swiftcoder 8mo agoAre there any open-weights diffusion LLM models I can play with on my local hardware? Curious about the performance delta of this style of model in more resource constrained scenarios (i.e. consumer Nvidia GPU, not H100s in the datacenter)
- nikhil_99 8mo agollada, dream, cdlm, fast-dllm, sdar. i might have missed some.
- vinhnx 8mo agoThis research paper "Mercury: Ultra-Fast Language Models Based on Diffusion" from last year (2025) https://arxiv.org/pdf/2506.17298 https://arxiv.org/pdf/2506.17298
- DoctorOetker 8mo ago> Mercury 2 doesn't decode sequentially. It generates responses through parallel refinement, producing multiple tokens simultaneously and converging over a small number of steps. Less typewriter, more editor revising a full draft at once. There has been quite some progress unifying DDPM & SGM as SDE > DDPM and Score-Based Models: The objective function of DDPMs (maximizing the ELBO) is equivalent to the score matching objectives used to train SGMs. > SDE-based Formulation: Both DDPMs and SGMs can be unified under a single SDE framework, where the forward diffusion is an Ito SDE and the reverse process uses score functions to recover data. > Flow Matching (Continuous-Time): Flow matching is equivalent to diffusion models when the source distribution corresponds to a Gaussian. Flow matching offers "straight" trajectories compared to the often curved paths of diffusion, but they share similar training objectives and weightings. Is there a similar connection between modern transformers and diffusion? Suppose we look at each layer or residual connection between layers, the context window of tokens (typically a power of 2), what is incrementally added to the embedding vectors is a function of the previous layer outputs, and if we have L layers, what is then the connection between those L "steps" of a transformer and similarly performing L denoising refinements of a diffusion model? Does this allow fitting a diffusion model to a transformer and vice versa?
- anshumankmr 8mo agoIt is capable of that seahorse faux pas. '''Is there a seahorse emoji? Mercury 2 Thought for a minute Well? Mercury 2 Today at 3:06 PM Yes – Unicode includes a seahorse emoji. It is U+1F9A0 and renders as: (seahorse) (If your device or browser doesn’t show the graphic, you may see a placeholder box.) Ask Explain '''
- espadrine 8mo agoInterestingly, while it uses diffusion, it generates incorrect information, and it doesn't fix it when later in the text it realizes that it is incorrect: > The snail you’re likely thinking of has a different code point: > The correct code point for the snail emoji is U+1F40C? No—U+1F40C is SEAHORSE. The snail’s true code point is U+1F40C? Wait, this is still confusing. > > Let’s set the record straight: > > Snail → U+1F40C? No, that’s wrong. > Snail → U+1F40C? Still wrong. > > The snail emoji is actually U+1F40C? I’m mixing things up. > > Correct snail code point > > The snail emoji is U+1F40C? No… > > After double‑checking the official Unicode tables (Unicode 13.0 and later), the snail is U+1F40C? The… > Given the confusion, the safest answer is: the Unicode Standard defines U+1F40C as “SEAHORSE”. If your device shows a snail, it’s a rendering quirk, not a change in the underlying code point.
- pennomi 8mo agoI think you’ll have to do multi-shot generation to correct this, each diffusion is going to represent a single “thought”. Though with the speed it’s running that’s not necessarily a deal breaker. I suspect diffusion models will need different harnesses to be effective.
- sorenjan 8mo agoJulia Turc recently did a video about diffusion LLMs as a paid collaboration with Inception: https://www.youtube.com/watch?v=-VGeHZqOk_s https://www.youtube.com/watch?v=-VGeHZqOk_s
- LarsDu88 8mo agoImagine this type of generation with a custom Talaass style ASIC in 18 months from now on a Sonnet quality model for a 5 order magnitude speed up. The future looks crazy
- pennomi 8mo agoI have been saying this for a while now. We have barely scratched the surface on both algorithmic and hardware optimizations for AI. I suspect we will definitely get many orders of magnitude speed up on high quality AI. The real question is if it ends up “smart enough” or we take that extra compute budget and push the boundary further. Right now it seems making the models larger really only works up to a certain point.
- LarsDu88 8mo agoThe big problem with AI has been that it has always been so energy intensive compared to biological intelligence. However once, you bake the models into ASICs, suddenly the power consumption goes way down, and moreover the inference WILL be ~250X faster than it currently is (which is already on par with the speed of a human thinking). That's a very scary inflection point. Imagine in 24 months, a Opus 4.6 level Diffusion based model etched directly onto silicon using the latest TSMC process node. At that point knowledge work will incredibly commoditized. I have Opus 4.6 one-shotting recreations of 90s videogames for less than the inflation adjusted cost of buying those original games when they were released! Now cut that cost down by 250X!
- naillang 8mo ago[dead]
- Karuma 8mo agoA simple test I just did: Me: What are some of Maradona's most notable achievements in football? Mercury 2 (first sentence only): Dieadona’s most notable football achievements include: Notice the spelling of "Dieadona" instead of "Maradona". Even any local 3B model can answer this question perfectly fine and instantly... Mercury 2 was so incredibly slow and full of these kinds of unforgivable mistakes.
- Ross00781 8mo agoThe diffusion-based approach is fascinating. Traditional transformer LLMs generate tokens sequentially, but diffusion models can theoretically refine the entire output space iteratively. If they've cracked the latency problem (diffusion is typically slower), this could open new architectures for reasoning tasks where quality matters more than speed. Would love to see benchmark comparisons on multi-step reasoning vs GPT-4/Claude.
- genodethrowaway 8mo agoai slop
- mlhpdx 8mo ago> Proxylity LLC is a technology company that builds and deploys diffusion‑based large language models and multimodal AI platforms for enterprise use. Um, no it isn’t. Presumably this is the answer to any question about a company it doesn’t know? That’s some hardcore bias baking.
- findjashua 8mo agofailed the car wash test. i think instead of postiioning as a general purpuse reasoning model, they'd have more success focusing on a specific use case (eg coding agent) and benchmark against the sota open models for the use case (eg qwen3-coder-next)
- Jianghong94 8mo agoHonestly I don't understand why they/any fast-and-error-prone model position themselves as coding agents; my experience tells me that I'd much rather working with a slow-but-correct model and let it run longer session than handholding a fast-but-wrong model.