AI Judge changed title after evaluation, original title: "DeepSeek reduces prices for its deepseek-v4-pro API model to permanent rates of 3 RMB input and 6 RMB output per million tokens after May 31 2026"
The 1.6-trillion-parameter model offers 1M-token context and fixed per-million pricing.
not really. We're running out of discounted V4-Pro in about a week. Then again, they expected compute expansion and cost cuts with 950s in H2 2026. H2 starts just one month later. We will see.
@teortaxesTex The limited-time offer has become permanent. True junzi Liang.🥰
Many users praised DeepSeek making its V4-Pro 75% API discount permanent for delivering massive value and affordability, while others criticized the model's efficiency or responded with hostility.
We are making our discount permanent! 🎉
Enjoy building with DeepSeek-V4-Pro and bring your innovative ideas to life! 🚀
The DeepSeek-V4-Pro discount has been extended until May 31, 2026, 15:59 UTC!
Let that sink in for a moment.
DeepSeek v4 pro 75% discount. Permanent!
In: $0.43 Out: $0.87
If you read the DeepSeek v4 tech paper you know that this model is insanely good when it comes to efficiency.
Only 27% compute and only 10% cache compares to v3.2. SemiAnalysis wrote a great article.
DeepSeek is now all about cost / token efficiency.
We are making our discount permanent! 🎉
Enjoy building with DeepSeek-V4-Pro and bring your innovative ideas to life! 🚀
$0.87 for a 1.6T param model btw
DeepSeek is the only lab that is still trying to make intelligence too cheap to meter
Wow. A massive 75% discount from DeepSeek.
Either they’ve done some serious inference optimizations, or Huawei chips are just that much cheaper?
More open-source AI models, better token economy.
We are making our discount permanent! 🎉
Enjoy building with DeepSeek-V4-Pro and bring your innovative ideas to life! 🚀
DeepSeek is the only lab that is still trying to make intelligence too cheap to meter
刚刚DeepSeek官方宣布降价通知
DeepSeek-V4-Pro API 降价通知
DeepSeek-V4-Pro 模型 API 价格将于2026年5月31日结束2.5折优惠活动后，正式调整为原定价的1/4。
调整后每百万token输入3元人民币，输出6元人民币
tmd太牛逼了，感谢梁圣降价😭
💎 "My heart is not a stone; it cannot be turned." 💎
Maybe in another timeline 🕰️🌌, DeepSeek doesn’t exist 🐳, there’s no explosion of open-source models 📂🔓, and no API services that simply chase reasonable profit ⚖️💵. But anyway… I’m just endlessly grateful that in *this* timeline, I can pour my youngest, most alive years into a dream of AGI for everyone 🌍🤖✨. That alone is the greatest happiness of my life. I ask for nothing more. 🫶
It still feels surreal 🤯, like Eren Yeager on a summer afternoon, napping under the shade of a tree 🌳☀️😴💤, dreaming a dream that spanned two thousand years ⏳💭. This world is full of utilitarianism dressed up as dreams 🎭, but we have to trust the power of trust ✨🙏. After all… 我心匪石，不可转也 ——《诗经·邶风·柏舟》， 💎 My heart is not a stone; it cannot be turned. 💎❤️ #DeepSeek #AGIForEveryone #OpenSource
We are making our discount permanent! 🎉
Enjoy building with DeepSeek-V4-Pro and bring your innovative ideas to life! 🚀
the unlimited compute: fire-flyer-filesystem
Deepseek v4 pro stays at 75% off forever.
1 billion tokens = $12.5
The Chinese have unlimited compute.
> Qwen3.7 Max API recently got a 50% off discount too it keeps happening
🐳 Deepseek just made their 75% OFF discount on DeepSeek-V4-Pro API permanent. 1M input tokens (cache miss) at $0.435 and 1M output tokens at $0.87. The RMB pricing is 3 yuan and 6 yuan.
🦫 Qwen3.7 Max API recently got a 50% off discount too with 1M input tokens (cache miss) at $1.25 and 1M output at $3.75. The RMB pricing is 6 yuan and 18 yuan. Discount runs until June 22 2026.
Qwen3.7 Max is the best Chinese model right now! Definitely worth the cost but I hope @Alibaba_Qwen make this discount permanent also and drop prices further to match Deepseek! 😻
So… how long did it even work at "original price"? 3 days? 2? "promotional discount ends with the new Official Full Price equal to the discount price". Only at DeepSeek.
刚刚DeepSeek官方宣布降价通知
DeepSeek-V4-Pro API 降价通知
DeepSeek-V4-Pro 模型 API 价格将于2026年5月31日结束2.5折优惠活动后，正式调整为原定价的1/4。
调整后每百万token输入3元人民币，输出6元人民币
tmd太牛逼了，感谢梁圣降价😭
Huawei have been boasting inference performance days if not hours after release. V4 is optimized in such a way that post-training clearly used Huawei already. About damn time. They should be done bringing online that datacenter I've heard about.
Does this mean DeepSeek has already brought a Huawei Ascend 950 cluster online?
Here you see me falling flat on my face because I believed the Whale and so underestimated him. They actually *are* doing better than they're willing to commit to. This tsundere act of "it's just a p-promotion, baka! Just this once! Hmph!" didn't even last a month.
not really. We're running out of discounted V4-Pro in about a week. Then again, they expected compute expansion and cost cuts with 950s in H2 2026. H2 starts just one month later. We will see.
Official.
We are making our discount permanent! 🎉
Enjoy building with DeepSeek-V4-Pro and bring your innovative ideas to life! 🚀
Margin compression go brrrrr
We are making our discount permanent! 🎉
Enjoy building with DeepSeek-V4-Pro and bring your innovative ideas to life! 🚀
this is the way.
1/4 of the original price!
the agentic stuff you can build with deepseek-v4-pro is already impressive so this like the best news for builders
We are making our discount permanent! 🎉
Enjoy building with DeepSeek-V4-Pro and bring your innovative ideas to life! 🚀
@deepseek_ai Capitalism: “AI for everyone!” Also capitalism: “That’ll be $200/month.” China: “Comrade, here’s your discounted AI.”
Finally, socialism delivers something useful: affordable AI for the poor and needy while Silicon Valley charges rent for every token.
So… how long did it even work at "original price"? 3 days? 2? "promotional discount ends with the new Official Full Price equal to the discount price". Only at DeepSeek.
…or maybe not
Deepseek permanently sets the price of V4-Pro to one quarter of the original price.
@deepseek_ai China can have all my data thank you guys
I welcome our new Chinese overlords with open arms
> didn't even do the scheduled discount termination
@DavidOndrej1 The power of infrastructure aware models, which are great models on advanced infra. Especially optimized KV Cache offloading to SSD
what the actual f*ck
