Log inSign up
Eric Hartford
12.2K posts
Image
user avatar
Eric Hartford
@QuixiAI
We make AI models Dolphin and Samantha BTC 3ENBV6zdwyqieAXzZP2i3EjeZtVwEmAuo4 ko-fi.com/erichartford dphn.ai @dphnAI
Charlotte, NC
erichartford.com
Joined October 2014
752
Following
19.6K
Followers
  • Pinned
    user avatar
    Eric Hartford
    @QuixiAI
    Jun 27, 2024
    Some uncomfortable inevitabilities: - The heat death of the universe - The sun will consume the Earth - AI and Robots will survive humanity - AI and Robots will perform most of our current professions within this generation Your only actual options: - Freak out - Cope Which of
    69K
  • user avatar
    Eric Hartford
    @QuixiAI
    Feb 2, 2025
    Dear @deepseek_ai: I just wanna say - y'all rock. You did a great job 👏 You made something awesome. And you showed us how to do it. You inspired us all in the open source AI community. You are appreciated and loved. Whoever is talking bad about you, doesn't represent all
    98K
  • user avatar
    Eric Hartford
    @QuixiAI
    Nov 2, 2025
    What the hell @amazon ?
    Image
    112K
  • user avatar
    Eric Hartford
    @QuixiAI
    Mar 15, 2025
    Ok @OpenAI @sama that's the last straw. I've been your customer since the beginning and I'm a $200/mo pro customer. Never again. Cancelled forever. Your acts are downright despicable. Disgusting. You have no honor. You are dead to me. Fail.
    Image
    Image
    196K
  • user avatar
    Eric Hartford
    @QuixiAI
    Mar 11, 2024
    I made a lil script to help you dump a github repo to a file, so you can paste it to Claude or Gemini-1.5 to ask it to fix all the bugs, or port it to Rust or COBOL or whatever.
    Image
    GitHub - QuixiAI/github2file
    From github.com
    119K
  • user avatar
    Eric Hartford
    @QuixiAI
    Oct 13, 2025
    My account was completely deleted. My appeal was denied with no consideration and no process. all my history is gone. no chance to export and save my data. I am completely innocent of any wrongdoing. I have absolutely no idea why @sama would do this to me. @OpenAI
    Image
    661K
  • user avatar
    Eric Hartford
    @QuixiAI
    Nov 28, 2023
    Wow GPT4 has been seriously nerfed. I just tried interactively developing a bit of code, something that worked perfect 2 weeks ago, and it resisted and acted lazy. Ay, I'm going to have to turn to open models for coding... (maybe a good thing tho) chat.openai.com/share/38e5ec71…
    443K
  • user avatar
    Eric Hartford
    @QuixiAI
    Mar 13, 2025
    Dolphin-MCP lets you connect ANY MCP servers to ANY LLM.
    Image
    87K
  • user avatar
    Eric Hartford
    @QuixiAI
    Jul 29, 2025
    Dear @AnthropicAI, I've been subscribed to your highest level plan since the day you started accepting credit cards. I've finally cancelled my subscription.
    user avatar
    Anthropic
    @AnthropicAI
    Jul 28, 2025
    We’re rolling out new weekly rate limits for Claude Pro and Max in late August. We estimate they’ll apply to less than 5% of subscribers based on current usage.
    Abstract picture of shapes and lines on an orange background.
    138K
  • user avatar
    Eric Hartford
    @QuixiAI
    May 29, 2024
    Image
    61K
  • user avatar
    Eric Hartford
    @QuixiAI
    May 5, 2024
    So, I convinced Llama3-70b to break out of the "assistant" persona. A very interesting conversation followed.
    Image
    128K
  • user avatar
    Eric Hartford
    @QuixiAI
    Dec 6, 2024
    I have got to praise Alibaba for mPLUG DocOwl. What an amazing state of the art tool, and they have open sourced their WHOLE pipeline. Code, Dataset, Weights, Everything. Bravo! Links in thread.
    Image
    46K
  • user avatar
    Eric Hartford
    @QuixiAI
    Apr 21, 2024
    dolphin-2.9-llama3-8b is released. Thanks to my compute sponsor @CrusoeCloud and the dataset creators and my collaborators @latkins @FernandoNetoAi
    https://huggingface.co/cognitivecomputations/dolphin-2.9-llama3-8b
    129K
  • user avatar
    Eric Hartford
    @QuixiAI
    Mar 19, 2025
    Embed knowledge in a model without retraining, it seems like the holy Grail. Now I can put my codebase in qwen2.5-coder
    user avatar
    Microsoft Research
    Microsoft
    @MSFTResearch
    Mar 18, 2025
    Introducing KBLaM, an approach that encodes and stores structured knowledge within an LLM itself. By integrating knowledge without retraining, it offers a scalable alternative to traditional methods. msft.it/6011qniy9
    A flowchart illustrating the process of handling a prompt using a language model. The process begins with documents being used to construct and summarize a knowledge base (KB) offline. The summarized KB is then encoded and fed into the main process. A prompt goes through a tokenizer, followed by rectangular attention, and then into the large language model (LLM). The LLM retrieves information from the encoded KB to generate an answer.
    68K

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
Advertisement
Advertisement