I don’t even understand half the words written there. Makes e happy.
If anyone is interested, basically Anthropic’s Claude has been leading the charge for code generation in the generative AI field for a while now.
Their latest model, Claude Fable, was topping charts and Anthropic was charging significantly more money to use it.
Soon after, OpenAI’s ChatGPT and Moonshot AI’s Kimi received updates (ChatGPT 5.6 and Kimi K3) that offered similar capabilities to Claude Fable however at a much lower price point.
The image shared shows Anthropic increasing usage limits of Claude Fable and also giving out free credit to customers as ChatGPT 5.6 and Kimi K3 released.
That is the best unbiased explanation of the events I can give.
Based on what I’ve read currently ChatGPT5.6 is the cheapest of the three, followed by Kimi K3 and finally Claude Fable. It’s kind of hard to do a direct price comparison just because of the number of factors that you need to take into account.
Now for my biased takes:
- I think Anthropic is trying to become profitable which is why they tried to sell Fable at such a high price. They tried to use their lead in the field to justify these prices.
- I think OpenAI is selling ChatGPT 5.6 at a huge loss just to capture the market which is why it’s so cheap.
- Not sure about Kimi’s position here. I have not been keeping up with Moonshot as much but it sounds like Kimi K3’s price is not being subsidized and it’s much closer to selling the product at cost. If true, this is very bad news for Anthropic.
Anyways I work in tech so I’m forced to be up to date on it :/
Anyways I work in tech so I’m forced to be up to date on it :/
I didn’t understand the post either. I appreciate your sacrifice. 🫡
I understand everything I have read here, but I fail in seeing something funny or ironic.
The humour is just in the way their tune has changed as the open source models have caught up. From “this is super exclusive, only for our favourites with the most money, and even then we’re limiting you” to “have some free credits! Please, try it!”. It’s always enjoyable to see a big company panicking a little in the face of open source alternatives.
To be fair, competition in business does feel like a novelty these days.
The whole point of OpenAI and Anthropic is that only the USA has the resources to train frontier models. The fact that China, under massive sanctions on chips, in training models on par with OpenAI and Anthropic (many of which are released with open weights) is really bad news for the AI bubble.
I think it’s pretty well known that a lot or all of the Chinese AI models are trained on ChatGPT/Claude so to be fair to OpenAI/Anthropic, they are probably the two companies that are mostly behind pushing AI capabilities forwards at the moment. It’s not 100% proven but there is a lot of evidence. That being said, I don’t feel bad for OpenAI/Anthropic personally due to how the original models were trained.
Not really, that’s just the data side of the equation. Both kimi and glm used deepseek’s architechture, deepseek has been kinda frontier in efficient/cost effect intelligence (acc to artificial analysis).
So the answer is much more nuanced than “they just copied chatgpt”.
Yeah I don’t think they just copied the other models end of story. If that was true it would be impossible for Kimi K3 to be in the same realm as Fable/GPT-5.6 since they all released around the same time. The Chinese companies are clearly also bringing new things to the table. But I think it’s highly likely most of the Chinese models have been at one point trained on at least one of the US models.
Again I don’t care that they did this.
I think Kimi is mainly a Chinese Frontier Lab working to advance the field and establish themselves as well as capture the market. I also think Kimi K3 will be much cheaper through third parties one the Open Weights drop. Its funny seeing Anthropic getting shit on like this after all the PR stuff they went through with releasing Fable (then Mythos) for corporate entities first because it was supposedly “too dangerous” to be in public hand.
Is ChatGPT still the worst of the three? We use it at work and Chad is an idiot.
Surprisingly they are now the best for the money, mostly because they are heavily subsidizing GPT-5.6.
Fable 5 as in the game?
Did they skip Fable 4?
The original post is titled “GLM/Kimi effect”. GLM models are developed by another Chinese company Z.ai. Their latest model GLM-5.2 has been benchmarked at similar coding capability as Anthropic’s Fable 5. You can use GLM-5.2 free via chat.z.ai. Like any of the “free” offerings it is limited in context, and be sure you don’t share any personal information with it because they will use your data for training (and who knows what else). But compared to other free models (eg those available via duck.ai) it is pretty impressive.
Also GLM-5.2 is open source, so that will make Dario’s head explode too!
Sorry to be pedantic, but I think it’s important not to call these models “open source”. Open weight is the equivalent of a compiled binary that can be run locally, so you’re basically getting Windows where it’s legal to reverse engineer, not Linux. The weights are permissively licensed, so the community is still free to quantize, fine-tune, reap, etc., which is analogous to something like Tiny11. Without the training data and the pipeline code, though, if Alibaba or Moonshot decide to go closed, the community won’t be able to fork their models and meaningfully continue development.
It’s better than what we are getting from “OpenAI” or Anthropic for the moment, but these are Chinese, for-profit companies that are no more altruistic than Google or Meta, and their open weight LLMs are just as likely to go closed as Llama or Gemma if the companies deem that to be in their best interest.
Fair enough. No one was saying they are altruistic 😉
Oh man this is funny
Especially if you know what the actual word means. In an infinite synced linked ST copyverses, Picard doesn’t have enough to do the move properly about that. There was this comic about a spacesuit person holding his helmet and see themselves on the top of the helmet doing the same thing. I can’t remember the line. I broke that tower a while ago and I can’t remember why…
