Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Mercury: Commercial-scale diffusion language model

inceptionlabs.ai

Post reply on HN
Print view View original

81–90 of 189 posts

Re: Mercury: Commercial-scale diffusion language model

#81

Post by fc417fc802 »

adrianmonk wrote: post #73
firejake308 wrote: post #51

Earlier quoted context omitted.

The rate of change of temperature is proportional to the difference in temperature. Adding the cold milk later let's you juice a larger temperature differential out of the 2-minute waiting phase. So you should add the milk later.

> the cold milk later I think what the other person is asking about is: can you be sure that the milk is (as) cold later? There's a lot of discussion about what happens to the temperature of the coffee during the 2 minutes. What happens to the temperature of the milk during that same time? Where is the milk stored? Do you grab it out of the refrigerator the moment you add it to the coffee? Or the cold milk sitting ou…

It's a safe bet that freshly brewed coffee is much farther from room temperature than refrigerated milk is. However deriving properties related to that symmetry (or lack thereof) would make an excellent question for an exam in an introductory class.

Re: Mercury: Commercial-scale diffusion language model

#82

Post by klipt »

fizx wrote: post #80
XenophileJKO wrote: post #29

Earlier quoted context omitted.

So my personal belief is that diffusion models will enable higher degrees of accuracy. This is because unlike an auto-regressive model it can adjust a whole block of tokens when it encounters some kind of disjunction. Think of the old example where an auto regressive model would output: "There are 2 possibilities.." before it really enumerated them. Often the model has trouble overcoming the bias and will hallucinate…

Once that auto-regressive model goes deep enough (or uses "reasoning"), it actually has modeled what possibilities exist by the time it's said "There are 2 possibilities.." We're long past that point of model complexity.

But as everyone knows, computer science has two hard problems: naming things, cache invalidation, and off by one errors.

Re: Mercury: Commercial-scale diffusion language model

#83

Post by danenania »

tyre wrote: post #19
inerte wrote: post #7

Not sure if I would tradeoff speed for accuracy. Yes, it's incredible boring to wait for the AI Agents in IDEs to finish their job. I get distracted and open YouTube. Once I gave a prompt so big and complex to Cline it spent 2 straight hours writing code. But after these 2 hours I spent 16 more tweaking and fixing all the stuff that wasn't working. I now realize I should have done things incrementally even when I hav…

Check out RooCode if you haven’t. There’s an orchestrator mode that can start with a big model to come up with a plan and break down, then spin out small tasks to smaller models for scoped implementation.

If you’re open to a terminal-based approach, this is exactly what my project Plandex[1] focuses on—breaking up and completing large tasks step by step.

1 - https://github.com/plandex-ai/plandex

Re: Mercury: Commercial-scale diffusion language model

#84

Post by emmelaich »

jonplackett wrote: post #5

Ok. My go to puzzle is this: You have 2 minutes to cool down a cup of coffee to the lowest temp you can You have two options: 1. Add cold milk immediately, then let it sit for 2 mins. 2. Let it sit for 2 mins, then add the cold milk. Which one cools the coffee to the lowest temperature and why? And Mercury gets this right - while as of right now ChatGPT 4o get it wrong. So that’s pretty impressive.

I had it write a Python program to calculate disk usage by directory -- basically a `du` clone. It was astonishly fast (2s) and correct. I've tried other models which have got it wrong, slow, and they've ignored my instructions to use topdown=False in the call to walk().

Re: Mercury: Commercial-scale diffusion language model

#85

Post by bayesianbot »

jtonz wrote: post #40

I would be interested to see how people would apply this working as a coding assistant. For me, its application in solutioning seem very strong, particularly vibe coding, and potentially agentic coding. One of my main gripes with LLM-assisted coding is that for me to get the output which catches all scenarios I envision takes multiple attempts in refining my prompt requiring regeneration of the output. Iterations are…

I basically did this with aider and Gemini 2.5 few days ago and was blown away. Basically talked about the project structure, let it write the final plan to file CONVENTIONS.md that gets automatically attached to the context, then kept asking "What should we do next" until tests were ready, and then I just ran a loop where it modifies the code and I press Return to run the tests and add the output to prompt and let it go again.

About 10 000 lines of code, and I only intervened a few times, to revert few commits and once to cut a big file to smaller ones so we could tackle the problems one by one.

I did not expect LLMs to be able to do this so soon. But I just commented to say about aider - the iteration loop really was mostly me pressing return. Especially in the navigator mode PR, as it automatically looked up the correct files to attach to the context

Re: Mercury: Commercial-scale diffusion language model

#86

Post by djrj477dhsnv »

maytc wrote: post #17
jonplackett wrote: post #5

Ok. My go to puzzle is this: You have 2 minutes to cool down a cup of coffee to the lowest temp you can You have two options: 1. Add cold milk immediately, then let it sit for 2 mins. 2. Let it sit for 2 mins, then add the cold milk. Which one cools the coffee to the lowest temperature and why? And Mercury gets this right - while as of right now ChatGPT 4o get it wrong. So that’s pretty impressive.

That example is probably in the training data? The puzzle assumes that the room temperature is greater than the cold milk's temperature. When I added that the room temperature is, say, -10 °C, Mercury fails to see the difference.

Under any reasonable assumptions for the size and shape of the cup, the amount of coffee, the makeup of the air, etc., the room being -10c won't change the result.

It would only matter if the air were able to cool the coffee to a temperature less than that of the milk in under 2 minutes.

Re: Mercury: Commercial-scale diffusion language model

#87

Post by FilosofumRex »

fc417fc802 wrote: post #81
adrianmonk wrote: post #73

Earlier quoted context omitted.

> the cold milk later I think what the other person is asking about is: can you be sure that the milk is (as) cold later? There's a lot of discussion about what happens to the temperature of the coffee during the 2 minutes. What happens to the temperature of the milk during that same time? Where is the milk stored? Do you grab it out of the refrigerator the moment you add it to the coffee? Or the cold milk sitting ou…

It's a safe bet that freshly brewed coffee is much farther from room temperature than refrigerated milk is. However deriving properties related to that symmetry (or lack thereof) would make an excellent question for an exam in an introductory class.

The two options are equivalent, since the final (equilibrium) temp of an adiabatic system (coffee + Milk + room) must be the same - ie it's the total amount of heat transferred that matters, and not the rates of heat transfer.

If the system is not adiabatic, ie the room is not big enough to remain constant temp, or equilibrium is not achieved in 2 mins of cooling, then the puzzle statement must be specify all three initial temps to be well poised.

Re: Mercury: Commercial-scale diffusion language model

#88

Post by NitpickLawyer »

jbellis wrote: post #61
jph00 wrote: post #25

The linked page only compares to very old and very small models. But the pricing is higher even than the latest Gemini Flash 2.5 model, which performs far better than anything they compare to.

Sort of. The benchmarks showing Flash 2.5 doing really well are benchmarking its thinking mode, which is 4x more expensive than Mercury here

Is cost really the main differentiator here, tho? "Solving" coding seems like the holy grail atm (and I agree, it can enable a bunch of things once that's done) and "traditional, organic, human fed code" is pretty expensive atm, so does cost really matter now?

Put another way, how much would company x be willing to spend on "here's a repo, here are the tests, here is the speed now, make this faster while still passing all the tests". If it "solves" something in cudnn that makes it 10% faster, how much would nvidia pay for this? 1m$? 10m$?

Re: Mercury: Commercial-scale diffusion language model

#89

Post by fc417fc802 »

FilosofumRex wrote: post #87
fc417fc802 wrote: post #81

Earlier quoted context omitted.

It's a safe bet that freshly brewed coffee is much farther from room temperature than refrigerated milk is. However deriving properties related to that symmetry (or lack thereof) would make an excellent question for an exam in an introductory class.

The two options are equivalent, since the final (equilibrium) temp of an adiabatic system (coffee + Milk + room) must be the same - ie it's the total amount of heat transferred that matters, and not the rates of heat transfer. If the system is not adiabatic, ie the room is not big enough to remain constant temp, or equilibrium is not achieved in 2 mins of cooling, then the puzzle statement must be specify all three i…

What gave you the idea that this is an adiabatic system? It's a cup of freshly brewed coffee on your kitchen counter. Equilibrium will certainly not be achieved within 2 minutes. Even if it were, different schemes would reach it at different time points.

It is fundamentally a question about rate of energy transfer.

The thing to notice about the symmetric system is that both items will experience the same rate of transfer. However there's presumably more coffee in your coffee than there is milk so it's not actually symmetric. If you adjust the parameters for volume and specific heat to make the final mixed product symmetric then it no longer matters when you do the mixing.

Re: Mercury: Commercial-scale diffusion language model

#90

Post by dvdhs »

schappim wrote: post #41

It's nice to see a team doing something different. The cost[1] is US$1.00 per million output tokens and US$0.25 per million input tokens. By comparison, Gemini 2.5 Flash Preview charges US$0.15 per million tokens for text input and $0.60 (non-thinking) output[2]. Hmmm... at those prices they need to focus on markets where speed is especially important, eg high-frequency trading, transcription/translation services and…

Not sure how HFTs are relevant here
Post reply on HN