• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    DeepSeek Plans Significant API Price Increase

    Developers on X react to DeepSeek's notice of upcoming API price changes.

    DR
    TC
    KA
    21 Sources, 56d ago, first seen 56d ago

    TLDR

    DeepSeek announced it will raise prices for its API services substantially in the near future and told users to plan their usage accordingly. The notice followed the release of its V4-Flash-0731 model and a period of heavy demand. Developers quoted in the posts described prior rates as the lowest on the market and noted the change ends that era. Some replies observed the increase would still leave DeepSeek cheaper than rivals even after a fourfold hike. Others linked the move to curbing low-value usage while preserving the service's position.

    Combined views

    1.2M

    21 Sources, first seen 56d ago

    Combined views

    1.2M

    21 Sources, first seen 56d ago

    12.5K likes
    12.5K likes
    704 comments
    1.2K saves
    664 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    704 comments
    1.2K saves
    664 reposts
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    21 Sources

    @yacineMTBCrazy what I can do with a few tweets.. my bad
    @teortaxesTexOn one hand, they can afford this on brutal market logic alone. 0731 destroys every other Chinese model on price-performance (so long as you don’t need frontier intelligence or multimodal). They can hike 4x easily. But I don’t think Wenfeng likes this or thinks this way. They plainly can’t serve the demand. They expect that Pro GA will crush their inference capacity. They seem to reject subscription/plan model and Zhipu/Kimi-style lottery on principle. And they would rather use more compute for training, RL, research. So they might increase prices far, until demand goes down. Remember, they are under no pressure to generate returns, the money is locked for five years, and they cannot spend on more compute anyway because it’s more scarce than money. Wenfeng expects he won’t be able to spend over ¥20B this year. Dark days ahead.
    @firstadopterAnother Nvidia manufactured bear narrative bites the dust
    @ns123abc🚨 DeepSeek plans a SIGNIFICANT INCREASE in API prices soon >“please plan your usage accordingly” it’s over for the poors
    @xlr8harder@teortaxesTex Sadly this yields flash's Pareto frontier position to OpenAI
    @thdxr:)
    @delipraoOver the past week, despite heavy use with agents, I barely managed to spend more than $5 with the flash model. It was too good to be true for long.
    @zephyr_z9They don't have enough inference capacity to fulfil demand, so they are trying to raise prices and destroy demand One of the reasons why I'm not worried about closed-source lab revenue is that they or their investors (hyperscalers like Google, Amazon, Microsoft) control the majority of global inference capacity DeepSeek, Kimi, Zhipu (have less than 400MW of combined compute capacity) can't challenge Ant/OAI (have over 6GW of compute) The real threat for Ant/OAI is a company like Meta/xAI (have 3-4 GW of compute) starting a price war in the near term The long-term threat is on-prem adoption of open models with enterprises setting up mini clusters (Jensen wants this to happen as it reduces customer concentration risk)
    @TacoCohenRT @ns123abc: 🚨 DeepSeek plans a SIGNIFICANT INCREASE in API prices soon >“please plan your usage accordingly” it’s over for the poors ht…
    @rohanpaul_aiDeepSeek officially announced API pricing is going up “significantly” That the cheapest near-frontier model on the market, at roughly 3 cents a task, is about to cost a lot more. It landed a week after DeepSeek-V4-Flash-0731 shipped. Currently it ranks 2nd among open-weight models, behind only Moonshot AI's 2.8T-parameter Kimi K3. The rates today are 14 cents/28 cents per 1M input / output, about 105X cheaper per task than Claude Fable 5. Then demand broke the arrangement. The new checkpoint became the fastest growing model ever by token usage on Ollama, which is now adding capacity across the US and Europe. Serving that much traffic at a price built to prove a point stops working once everyone concedes it. But now, the timing of this price increase invites trouble, because Meta's Muse Spark and OpenAI's GPT-5.6 Luna now match it on both capability and price.

    21 Sources

    @yacineMTBCrazy what I can do with a few tweets.. my bad
    @teortaxesTexOn one hand, they can afford this on brutal market logic alone. 0731 destroys every other Chinese model on price-performance (so long as you don’t need frontier intelligence or multimodal). They can hike 4x easily. But I don’t think Wenfeng likes this or thinks this way. They plainly can’t serve the demand. They expect that Pro GA will crush their inference capacity. They seem to reject subscription/plan model and Zhipu/Kimi-style lottery on principle. And they would rather use more compute for training, RL, research. So they might increase prices far, until demand goes down. Remember, they are under no pressure to generate returns, the money is locked for five years, and they cannot spend on more compute anyway because it’s more scarce than money. Wenfeng expects he won’t be able to spend over ¥20B this year. Dark days ahead.
    @firstadopterAnother Nvidia manufactured bear narrative bites the dust
    @ns123abc🚨 DeepSeek plans a SIGNIFICANT INCREASE in API prices soon >“please plan your usage accordingly” it’s over for the poors
    @xlr8harder@teortaxesTex Sadly this yields flash's Pareto frontier position to OpenAI
    @thdxr:)
    @delipraoOver the past week, despite heavy use with agents, I barely managed to spend more than $5 with the flash model. It was too good to be true for long.
    @zephyr_z9They don't have enough inference capacity to fulfil demand, so they are trying to raise prices and destroy demand One of the reasons why I'm not worried about closed-source lab revenue is that they or their investors (hyperscalers like Google, Amazon, Microsoft) control the majority of global inference capacity DeepSeek, Kimi, Zhipu (have less than 400MW of combined compute capacity) can't challenge Ant/OAI (have over 6GW of compute) The real threat for Ant/OAI is a company like Meta/xAI (have 3-4 GW of compute) starting a price war in the near term The long-term threat is on-prem adoption of open models with enterprises setting up mini clusters (Jensen wants this to happen as it reduces customer concentration risk)
    @TacoCohenRT @ns123abc: 🚨 DeepSeek plans a SIGNIFICANT INCREASE in API prices soon >“please plan your usage accordingly” it’s over for the poors ht…
    @rohanpaul_aiDeepSeek officially announced API pricing is going up “significantly” That the cheapest near-frontier model on the market, at roughly 3 cents a task, is about to cost a lot more. It landed a week after DeepSeek-V4-Flash-0731 shipped. Currently it ranks 2nd among open-weight models, behind only Moonshot AI's 2.8T-parameter Kimi K3. The rates today are 14 cents/28 cents per 1M input / output, about 105X cheaper per task than Claude Fable 5. Then demand broke the arrangement. The new checkpoint became the fastest growing model ever by token usage on Ollama, which is now adding capacity across the US and Europe. Serving that much traffic at a price built to prove a point stops working once everyone concedes it. But now, the timing of this price increase invites trouble, because Meta's Muse Spark and OpenAI's GPT-5.6 Luna now match it on both capability and price.