DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

12 comments

• jiehong 5 minutes ago
  Sounds nice!

  But, the web ui chat version of flash has very poor language following abilities in my experience:

  You may ask it something in English, and get a thinking chain in Chinese with an answer in Chinese, or an English thinking chain and an English answer. Using the retry button on the same question has a 50/50 chance of any of those results.

  Sometimes, asking something in English, but where information are mostly in another language may make the answer in the language where data has been found. The other day, I asked something about a local German thing, in English, and I got an answer in German instead. It’s as if all the language data stirred it away from the language of the user’s question.

  [-]
  • swiftcoder 3 minutes ago
    I've hit this too, but you can just add "in English" to steer it
• postalcoder 16 minutes ago
  I hope DeepSeek takes some time to improve their tuning for reasoning effort. Right now, there are only three reasoning efforts: low, high, and max.

  For all intents and purposes, "low" is pretty much the same as turning reasoning off, and "high" is similar to "max". "High/max" performs way too much reasoning and takes forever. They really need a proper "medium" setting.

  I get it that they're probably focused on pushing performance right now, but the ergonomics of the model aren't great.
• oefrha 1 hour ago
  Source is apparently a banner announcement on https://platform.deepseek.com/usage. Had me searching for a couple minutes...
  [-]
  • nickweb 36 minutes ago
    I swear I put that at the start of the post. Must've managed to miss it when copy and pasting!
• EbNar 17 minutes ago
  Since a few months, I almost exclusively use the Chinese "flash" models for my need. They are a joy and they cost pennies per answer. Great job.
  [-]
  • darkoob12 0 minutes ago
    My mental bias always kept me away from Chinese models. Because i know that china is a surveillance state and all the things we know about CCP. But after what we learned about OpenAI and how they most likely used user data to basically cheat in an open competition i think it does not matter which AI provider you use all of them will own your data and all of them can spy on you. So I am willing to switch to Chinese models. This way we help them develop and improve models some day we can run them locally.
  • XzAeRosho 3 minutes ago
    Same for me. DeepSeek models are incredibly good at implementation and light planning. I still default to Opus models for feature planning, but for most simple features the Pro models suffice.

    Incredible good value and product they have built.
  • ActionHank 6 minutes ago
    I am legitimately more excited for this release than any frontier models at this point.

    I don't need a model that can invent new mathematics. I need something that is fast, cheap, and consistent. Give me that and I can build and scale.
• tarruda 40 minutes ago
  Hopefully it will be open weights and have the same architecture and size as the current v4 flash vision, which is probably the best LLM that can be run on 128G devices.
  [-]
  • fluoridation 27 minutes ago
    Interesting, I had assumed it'd be too large to fit. What quant and context size are you running?
    [-]
    • tarruda 6 minutes ago
      IQ3_XXS (~3.2 BPW). For me this is an option because my Mac studio is only used for serving LLMs, so I can afford to dedicate most of its RAM to this. I can run with 256k context and only uses ~117G, with the remaining (up to 125G which I can allocate to VRAM) being used for prompt caching and context checkpoints.

      I'm making my own quants, though the Vision-Exp version is outdated and won't work on llama.cpp master branch (I built it before llama added support):

      - https://huggingface.co/tarruda/DeepSeek-V4-Flash-0731-GGUF

      - https://huggingface.co/tarruda/DeepSeek-V4-Flash-Vision-Exp-...

      For the Vision-exp version, I also ran perplexity + KLD against the original MXFP4. Seems quite OK: https://huggingface.co/tarruda/DeepSeek-V4-Flash-Vision-Exp-...
• nickweb 33 minutes ago
  Via nitter: https://xcancel.com/JustinGorya/status/2097287080128708930

  Looks like the new model can be used if summoned via the API but the API won't list it.
• igleria 31 minutes ago
  v4 pro was decent then a better cheaper faster model comes now?

  As a consumer I feel like hansel and gretel combined, deepseek could be the witch.

  [-]
  • calgoo 9 minutes ago
    v4 flash has been working quite well for the majority of my personal projects, with occasional v4 pro or Kimi 3 for the most complicated tasks or to check the overall project progress (when vibe coding).
  • throwaway473825 19 minutes ago
    It's not unprecedented given that GLM 5.3 Flash was better and cheaper than GLM 5.2.
• swiftcoder 46 minutes ago
  If they can keep up this cadence of Flash leap-frogging the previous Pro, we're in for a good time
• nicce 22 minutes ago
  > In keeping with our commitment to user responsibility, following the official launch of V4.1 Flash and prior to the release of V4.1 Pro, all requests to the Pro model will be routed to V4.1 Flash and billed at Flash's price. If you encounter any issues during your comparative testing between V4 Pro and V4.1 Flash, please do not hesitate to reach out to us with your feedback. Thank you for your support!

  Wow. Imagine OpenAI/Google/Anthropic doing this! Nope.
• thrownaway561 19 minutes ago
  I will continue to be amazed by how much power you get from DeepSeek Flash for the cost. I have let that puppy lose on so many projects and it is has never let me down. It can build and entire Rails app in no time and even do the tests. For most things, I don't get why people pay the money for Claude. DeepSeek Flash is my default agent in Omarchy.
• neugls 1 hour ago
  Waiting to use it
• donk8r 22 minutes ago
  [flagged]