Hacker News

Ox-Alpha Is GLM?

67 points by jitbit ago | 41 comments

swiftcoder |next [-]

> How many words are in the previous message?

Its amazing to me that providers haven't added any sort of masking of the prompt in the thinking traces to avoid prompt extraction via this sort of trivial attack

|next |previous [-]

gvkhna |next |previous [-]

If it’s not zhipu then why is it returning errors that zhipu does for other models? Who else would return the exact same errors even if they took a lot of core infra like tokenizer from z?

ggcr |root |parent |next [-]

Ziphu has that many resources to be able to serve capacity for 1 quadrillion tokens per day on Nous portal? My bet is that it's a Composer model from Cursor running on xAI cluster, they already did a Composer based on Kimi-K2.5

e9 |root |parent |previous [-]

Someone could've trained model on top of GLM. Same way Cognition trained their SWE model on top of Kimi and Cursor did same with their Composer model.

gvkhna |root |parent |next [-]

While possible the amount of variation in serving infrastructure is unlikely to land with actually giving the exact same errors zhipu does.

It feels like glm flash, and there was a report zhipu had secured a huge new cluster suggesting they have the capacity. My guess anyway.

https://www.tomshardware.com/tech-industry/artificial-intell...

vitorgrs |root |parent |previous [-]

The reasoning levels are the same as GLM 5.3. GLM 5.3 is still not open...

I believe it's GLM 5.3 Flash or Air.

weiran |root |parent [-]

Reasoning levels are often just injected system prompts so not a great way to fingerprint models.

jerrythegerbil |next |previous [-]

As someone who uses NCD nearly every day, I have concerns about how it’s been used here.

But while we’re “guessing”: Xiaomi MiMO

walrus01 |root |parent [-]

Have also seen people guess it's a next version of Longcat, but I also think that's unlikely

tadkar |next |previous [-]

I wonder if the NCD metric says something about distillation too. Would you expect that a model that has been distilled/seen traces from other models would have a smaller NCD? It would be really interesting to see if this holds up and provides evidence of distillation or certainly evidence of model outputs being used in the training mix.

mogili |next |previous [-]

It's not a good model tbh, got a bunch of things wrong that Opus corrected in my codebase.

gvkhna |root |parent |next [-]

The harness is making a big difference, lackluster performance with pi but somehow very good performance with opencode. There’s some rl there for sure, for a smaller model it’s likely going to perform much better in a harness it understands the best.

petesergeant |root |parent |previous [-]

Yet to find a model that cross-model review doesn’t find a bunch of things wrong with. I’m running simultaneous review with whichever of Grok4.6/GLM5.3/Fable/Sol didn’t write it, and each model tends to find items the others didn’t.

mogili |root |parent |next [-]

Wasn't just a review, it failed the task I gave and Opus completed the task

epolanski |root |parent |previous [-]

If your changes are non trivial even the same model will loop over and over with the feedback.

volf_ |next |previous [-]

GLM 5.3 and all previous models don't have a vision encoder and can only accept text. Ox-Alpha can accept video and images, so unless Z-ai added a pretty good vision encoder for this model, I don't think so.

My money is on Moonshot and this being Kimi K3.5. The measured tps and latency is in-line with K3's tps and latency from Moonshot.

MiniMax M3.5 is also possible (but the MiniiMax provider is a lot more performant than the lab behind ox-alpha, so less likely).

nylonstrung |root |parent |next [-]

It would be stranger to me that Kimi switched to GLM's tokenizer than that GLM added multimodal like Kimi and Deepseek both did recently

minimaxir |root |parent |next |previous [-]

The other tell from the provider angle is capacity. Whoever is hosting Ox Alpha has a lot of capacity which narrows down a lot of the Chinese companies.

Bolwin |root |parent |next |previous [-]

Glm had made vision models in the past. Look up GLM 5v.

The only question now is if it's 5.3v, 5.4/5.5 or a dedicated flash/vision model

BoredomIsFun |root |parent |next [-]

GLM made pretty decent for that time small 9b vision model, GLM-4.1.

volf_ |root |parent |previous [-]

Yeah. It could be. The Z.ai DC latency is still ~1.2s faster than whomever is serving this model.

TiredOfLife |root |parent |next |previous [-]

But Moonshot limited signups because they lacked compute.

Almondsetat |root |parent |previous [-]

DeepSeek literally just came out with the vision-enabled version of Flash v4 which was purely text based. Why would GLM not be able to do the same thing?

volf_ |root |parent [-]

It's possible

try-working |next |previous [-]

Nvidia

xorgun |next |previous [-]

Dont rule out ssi

nullbio |root |parent [-]

That would be insanely disappointing.

petesergeant |next |previous [-]

I think within 12 months we’re going to see a frontier (inc open models) that’s so good at almost all human-directed tasks that which model you use just won’t matter. Only differences that remain will be in deep research or very long-range tasks.

stingraycharles |root |parent [-]

People were saying this last year, and they’ll be saying the exact same thing next year. The goalpost keeps moving.

Tepix |root |parent |next [-]

It‘s already happening, people are using cheaper models because they are good enough

petesergeant |root |parent |previous [-]

Someone else having been too early on a prediction has little bearing on my prediction. A year ago almost nobody was using open models as daily drivers, today they are. When I run out of Fable and Sol credits in a week, I switch to GLM5.3, and it's not quite there, but it's good enough for productive work.

ChrisArchitect |next |previous [-]

Related:

Ox Alpha

https://news.ycombinator.com/item?id=49381896

behnamoh |previous [-]

[flagged]

walrus01 |root |parent |next [-]

I wasn't aware that an inanimate piece of software run by a corporation can be 'doxxed'. Totally inaccurate use of the neologism.

minimaxir |root |parent |next |previous [-]

You cannot "dox" an AI model.

Given the traction the model has received, it is extremely newsworthy to know who's developing and hosting it.

behnamoh |root |parent [-]

my question is: how does that affect a company's strategy? it's not like management is gonna switch models soon as a new shiny one drops. entire workflows depend on specific models working the way they do; you can't just swap out models.

minimaxir |root |parent |next [-]

If it's a really really good model, then yes, people will switch as long as the price is right. Ox Alpha is looking to be a really really good model to the point that it competes with Fable/Sol, and will likely beat them on price.

volf_ |root |parent |previous [-]

same business model as crack. best way to get people hooked is to make the first hits free.

JSR_FDED |root |parent |next |previous [-]

You can learn from the variety of techniques they used to come to this conclusion.

tjwebbnorfolk |root |parent |previous [-]

> so much time on your hands

Yet here you are, reading AND commenting about it

behnamoh |root |parent [-]

and flagging it.