Yet another Hacker News

183 points theykk 2 hours ago 79 comments

freakynit 2 hours ago | parent

Those benchmarks can't possibly be real, right? Right? <insert padme 4-panel meme here>

logicchains 2 hours ago | parent

At this rate the frontier labs will be Anthropic and DeepSeek, with OpenAI struggling to keep up.

freakynit 2 hours ago | parent

I believe these are the last sets of super-high-end models available to public (open or closed).. these companies probably will now start to withhold top-of-the-line models for themselves, governments, and their super-high-paying partners only.

stymaar 2 hours ago | parent

News from a near future:

Here's How Distill APT Used McKinsey's Private Access To Top Anthropic Model To Help The Chinese Keep Up With The Us Top IA Labs Despite Access Restrictions

ReptileMan 1 hour ago | parent

We have the weights, we have the papers, nothing preventing development on this foundation.

A_D_E_P_T 1 hour ago | parent

That hasn't been the way to bet. DeepSeek and the other Chinese firms show no signs of slowing down, so if OAI and Anthropic halt releases and "pace the frontier" they'll just fall behind. (And there's a chance that high-end AI will become fully commoditized, which suits DeepSeek just fine.)

Besides, Fable 5.5 is reportedly releasing within the next few days, probably on Tuesday...

girvo 6 minutes ago | parent

China just announced tightened export controls on AI models, for what it's worth.

slopinthebag 1 hour ago | parent

idk, sol 6.1 is both cheap and capable. i've been using it exclusively, i don't need to care about usage, and it's really smart.

yes maybe opus 5.5 is stronger and flash is cheaper, but 6.1 strikes a good balance imo.

ReptileMan 1 hour ago | parent

i have a feeling that Sol is just Astra low, but with lowest possible priority on the compute.

WASDx 1 hour ago | parent

I believe it's Astra quantized. It was briefly named astra-lite.

computerex 16 minutes ago | parent

OpenAI models don’t come close to competing with deepseek in cost quality ratio. I have a $20 plus plan and I run out of my 5h limit in literal minutes, every tick of the harness is like 10% of my 5h limit.

You can in contrast get far more done with $20 worth of deepseek api spend. No comparison.

A_D_E_P_T 1 hour ago | parent

OpenAI's models -- and even Google's Argon -- have particular niches where they shine. For GPT-6/6.1, it's 3D modeling; Astra is hands-down the best at making .step files, Blender files, etc. For Argon, it's natural language writing and everything that hooks up to Google, e.g. search and maps-based itinerary planning.

Anthropic and DeepSeek (+ GLM) are definitely solving coding, though!

They're all quite different tools.

Kurtz79 1 hour ago | parent

Artificial Analysis recently released a dedicated 3D benchmark, where Astra lags behind both Opus and Sonnet 5.5 (and yes, open weight models seem to be quite far behind).

https://artificialanalysis.ai/articles/aa-3d

swiftcoder 56 minutes ago | parent

its unfortunate they didn't include deepseek in the initial batch of benchmarks

A_D_E_P_T 56 minutes ago | parent

Interesting!

For me it's large files in Blender (in some cases >400MB), mostly, where Astra excels. I gave it a look, but I don't know if the benchmark encompasses that sort of thing.

.step files are simpler pieces of straightforward code, so it makes sense that Opus and Sonnet 5.5 do very well...

bashtoni 2 hours ago | parent

"a minor update" they say before showing off benchmarks that display a substantial improvement.

jononor 2 hours ago | parent

Above GLM 5.3, nice! That is a very capable model. I do wish that we will see some smaller "flash" models again though - running this one locally is tough (very costly).

LaurensBER 2 hours ago | parent

For comparison, close to Opus 5. Far behind Sol 6.1 and Opus 5.5.

Probably a pretty decent model for 90% of the usecases!

I've been doing a lot of hobby work with GLM 5.3 Flash and I can count the amount of times that I need a stronger model on two hands.

I have access to unlimited tokens at works and Opus is great but also often overkill if you know exactly what you need.

egeozcan 1 hour ago | parent

I use Opus for planning, through API on medium thinking, it doesn't cost anything significant and the rest can be happily done with deepseek or glm flash. Using sol 6.1 for review loops at the end. All running on pi.

A very complex task tends to cost like 10$ max. It's comparable to a subscription with enormous flexibility.

It's a steal.

Pragmata 1 hour ago | parent

Cant you run it on a pair of DGX spark? Sure it's expensive now, but at the initial retail price that was what, 8k in hardware? Not exactly unachievable

bashtoni 1 hour ago | parent

No, this is not a small model. I believe 4x Sparks is possible but 4.1 is pretty big for a flash model.

MiMo 2.6 Flash is the most capable model you can currently run on 2x Sparks without resorting to a quantised version of something.

Pragmata 1 hour ago | parent

I meant quantized. You can get an EXL3 quantized version thats almost the same performance.

bashtoni 9 minutes ago | parent

My experience running quantised models is that they look the same at first but are all noticeably worse once the context size gets beyond anything trivial.

minraws 44 minutes ago | parent

By next year we will likely see GLM 5.3 level performance in Qwen 4 MoE models that can run on most higher end consumer systems.

I don't think there is a reason to care about any of it, just get a few dollars on a few openrouter or similar platforms and just use the models for now, wait for cheaper models that are good enough for you to become available for your hardware.

Sure you might never see frontier performance on consumer hardware but do you really need that though? Does anyone? And if you ever do just get a 30-day sub for that work, and cancel it.

m00dy 2 hours ago | parent

We're in a crisis

pmxi 2 hours ago | parent

Hilarious sentence: "DeepSeek-V4.1-Flash-1017 is a minor update to DeepSeek-V4.1-Flash, delivering substantial improvements across reasoning, coding, agentic tasks, and visual benchmarks." (emphasis mine)

freakynit 2 hours ago | parent

Kinda shows their team is more engineers than marketers.

stavros 17 minutes ago | parent

Strictly speaking, it just shows they aren't marketers. Bad copy doesn't tell us anything about their engineering skill.

nchmy 13 minutes ago | parent

Strictly speaking, they didn't say anything about their engineering skill

jaynetics 1 hour ago | parent

It's not really contradictory, minor changes can have major effects.

mdp2021 1 hour ago | parent

So they meant a "small" update - "conservative", "inexpensive"...

An example when using a less frequent term actually reduces specificity.

ReptileMan 1 hour ago | parent

That is how exponential functions work. Improve every aspect of a system with 1% and you may get easily 50% more performance.

cardboard9926 1 hour ago | parent

Hoping it's minor relative to what's coming

vrganj 1 hour ago | parent

It's about playing to two different audiences.

The CCP requires extensive certification for new major releases, to make sure a model doesn't do anything they don't want it to.

To get around this, Chinese labs only bump versions when they have to and usually claim it's a minor update.

Then on the other side, they're obviously trying to market to the tech crowd...

epolanski 30 minutes ago | parent

Source?

vrganj 28 minutes ago | parent

Drum Tower podcast (from the Economist, paywalled) about Chinese AI.

Would highly recommend it, it's IMO the only serious Western reporting about China.

surgical_fire 1 hour ago | parent

It's a bit refreshing to read updates from an AI lab without bullshit doomerism and overwhelming sense of self-importance from shit like OpenAI and Claude.

Also, DS is an excellent model.

nkmnz 31 minutes ago | parent

That's one small step for a man, one giant leap for mankind.

fivedogit 25 minutes ago | parent

I'm at 1 trying to get to 10. I move to 2. Minor update. 100% improvement towards goal.

These are not mutually exclusive.

throwfaraway135 2 hours ago | parent

Another big win for opensource AI.

BTW who would have thought China would be the champion of freedom and opensource in the AI arena, while US is fighting for corporate/government control.

Dah00n 1 hour ago | parent

Well, at least mostly everyone outside the US.

ReptileMan 1 hour ago | parent

Anyone that has observed China since the 90s.

There are different types of freedom and China's bet is that as long as they provide enough economic freedom the public will be ok with somewhat restricted social and political one.

InsideOutSanta 48 minutes ago | parent

Yeah, I had a somewhat odd conversation with a Chinese acquaintance last year when she got to talking about how America sees China. Her point was that, "How am I not free? I can do whatever I want, go where I want, buy what I want. I can talk to you and say whatever I want; nobody is here listening to me."

The fact that she can't call Xi "Winnie the Pooh" on Douyin wasn't something that occurred to her, or maybe that she was aware of at all. I didn't ask. I don't like to get into these discussions when I'm in China.

computerex 21 minutes ago | parent

There is no such thing as “free speech” anywhere. It’s all a spectrum and the Americans in particular are indoctrinated. It’s a red herring to distract from the real social issues. People in America have some of the worst healthcare in the world. But hey at least we have “free speech”.

ngl999 1 hour ago | parent

That's an image China wants the rest of the world to believe. Towards inside it still fits the typical stereotype.

npn 2 hours ago | parent

So this is the reason for the crash a few days ago.

dgellow 1 hour ago | parent

What crash?

brador 1 hour ago | parent

Ai/tech stock crash

npn 1 hour ago | parent

Deepseek servers went unavailable for 5 hours. They compensated that by rewarding the previous week usage as credit though. Pretty nice.

handfuloflight 2 hours ago | parent

Is it coming to OpenCode et al?

opengears 2 hours ago | parent

and do we know which version runs on opencode at the moment (on the free and on the paid version), is there a way to find out?

swiftcoder 1 hour ago | parent

The only way I'm aware of is to check their twitter feed. They moved to Deepseek v4.1 flash in September

kzrdude 1 hour ago | parent

Presumably. The current opencode slug for the model is just 'deepseek-flash' suggesting they will just update that one going forward, not keeping versions in parallel.

floooof 1 hour ago | parent

they must've been using this since 2-3 days when it became free again in OpenCode, it's noticeable smarter, beats GLM

nik736 1 hour ago | parent

Those are crazy numbers for a Flash model. Hopefully they can scale what they are doing to 4.1 Pro.

ytxm 1 hour ago | parent

It passed my threshold for not caring to try new models back with DS4, I use it for literally everything. Announcements of new models in the series always feel curious, because it doesn't really feel like there's much left to fix that would improve on what's already there for any task I know of.

It's a bit like obsessing over patch releases of some software after a major upgrade already solved the original problem long ago.

mmoustafa 1 hour ago | parent

definitely could use less thinking time and lower token usage, which is what this bump is focused on

xiphias2 1 hour ago | parent

I have quite the opposite experience: I can do any concrete coding change with any model, but the intelligence limit just doesn't go through more complex debugging examples.

And then when I try a smart model, like Astra medium and it solves it (I had a firmware bug in my new laptop where I couldn't charge through USB-C) on the first try (altough with a repo already having lots of information about the hardware/firmware/debugging programs/traces), I realize how much time I'm losing on the few examples that the less intelligent models are just not able to debug and solve.

KronisLV 1 hour ago | parent

Also they can work in completely new domains, like Astra and Opus 5.5 starting to be able to use something like Blender. Currently they still often do horrible topology and basic UV mistakes, and the animations are really bad, but who knows where they’ll be in 5 years.

squidbeak 25 minutes ago | parent

5 years? Look at the leaps since last November. Since the Spring, even.

ytxm 31 minutes ago | parent

It's not that more capable models aren't possible, it's about the day to day baseline and what is actually needed. Like carrying a workstation-class laptop and its two 500W power supplies everywhere when all you run is a version of Excel so old a mid 2000s netbook would suffice. This pretty much describes my typical computing experience. Galaxy brain models sound great and all, but the problems I have involve neither attempting to cure cancer nor any plan on attempting something like that in the foreseeable future.

mdp2021 1 hour ago | parent

> improve on what's already there for any task I know of

What about "good governance"?

Does it have zero hallucinations? Does it reflect thoroughly on every single idea it uses, verifying its solidity?

pimeys 1 hour ago | parent

Speed and price. You should try the 4.1 Flash on one of the fast providers. It's pretty crazy how fast this model works, and it's great for iterative software development because you can steer it fast to where you want to go.

https://www.lithosai.com/

https://www.coralbricks.ai/

swiftcoder 1 hour ago | parent

How are these high-speed providers close to matching the API rates from deepseek?

pimeys 1 hour ago | parent

Seems that V4.1 Flash is quite cheap to serve. Especially how it implements the cache, meaning the cache pricing can be very low depending on how you implement it in your inference engine.

Depends on your use case. But if you do software development with a harness such as oh-my-pi, your cache rates are probably somewhere above 98%. Coralbricks has free cache reads, meaning it's much cheaper than the Deepseek API rates.

Lithosai is more expensive especially in the ultra tier. It's if you want extremely snappy and fast responses, which is important e.g. for a personal assistant where every second counts.

If you generate a lot of output, you're better off with the Deepseek API rates.

andrepd 1 hour ago | parent

Where does one even get decent ZDR provider? I'm wary of sending sensitive data but not at a scale where I would build my own machine.

ytxm 56 minutes ago | parent

Scaleway have DS4. I don't think they call it ZDR, but their framework is very strong https://www.scaleway.com/en/docs/generative-apis/reference-c...

gmbhgmbh 1 hour ago | parent

Thank you deepseek, we are eating good today.

waximabbax 1 hour ago | parent

Looks very impressive especially considering the price and speed. Let's wait for AA to confirm how well their benchmarks hold up.

frail_figure 1 hour ago | parent

Why not just try it yourself?

viraptor 1 hour ago | parent

It costs money to run serious tests. We're way past the time when you could get an idea of the new model in a couple of tricky prompts. Now I need to compare a whole resulting project to get a clear view. If someone else proves the benchmarks don't hold up in reality, then it saves us further tests.

risphere 1 hour ago | parent

It's better than GLM 5.3 and Kimi K3 and way cheaper.

slopinthebag 1 hour ago | parent

that's honestly insane. it's like the go-to for when you don't want to use opus 5.5/sol 6.1

ngl999 1 hour ago | parent

The go-to is a local model.

ngl999 1 hour ago | parent

Censored, no surprise

vintermann 49 minutes ago | parent

I'd never heard of the BabyVision benchmark they do so well on. Pretty interesting stuff:

https://arxiv.org/abs/2601.06521

sharktheone 27 minutes ago | parent

beating K3 and GLM-5.3 is kinda insane