maxloh

Here's a beginner tutorial for the language: https://docs.toit.io/language/toitversus

Nice syntax. It looks like how we'd design programming if C had never existed.

  add a/int b/int -> int:
    return a + b
  
  main:
    print (add 2 3)  // Prints 5.

Yeah, I mentioned them in my comment. They're my favorite lab/vendor. Everything is open and reproducible.

It's just that their models were never frontier-level. By major, I meant frontier or popular.

Not really. That definition also requires the training code and training data composition to be open-sourced.

No major Chinese or Western labs satisfy these requirements. The only vendors I can think of that do are Nvidia and Allen AI.

I believe Meta's models were never truly frontier in the way Qwen, Kimi, etc., compare to Claude or ChatGPT.

Correct me if I am wrong.

We are now at the mercy of labs to provide frontier models as open-weight. Now that even some Chinese labs are releasing their models as weight-available, what stops them from prohibiting certain internal uses of new models? What about public chatbots?

Open models we have now will become obsolete within a few years. Imagine a world where our last open models were from the GPT-2 era.

I believe we need truly sovereign models, not in the Western sense, but from a community perspective. We need a model line that stays open-source across generations.

The OSI actually published an "open-source" definition for model weights, which requires the training code and training data composition to be published. The exact training corpus can remain private though.

https://opensource.org/ai/open-source-ai-definition

Calling this open weight is misleading. That's why we had huge backlash when Redis, Elasticsearch, and Terraform went source-available, right?

Weight-available is a better adjective for this kind of licensing. Same for Qwen, GLM, Kimi, MiniMax, etc. If they really want to be open source, they should pick an OSI-approved license.

> 2. If the Licensee or any of its affiliates operates a Model as a Service business, the Licensee must enter into a separate commercial license with Mistral before using the Model or its derivative works for any commercial purpose. The Licensee may contact Mistral, using this form: https://mistral.ai/contact/, to request a commercial license. [0]

Calling this "open weight" is misleading. That's why we had huge backlash when Redis, Elasticsearch, and Terraform went source-available, right?

Weight-available is a better adjective for this kind of licensing. Same for Qwen, GLM, Kimi, MiniMax, etc. If they really want to be open source, they should pick an OSI-approved license.

[0]: https://huggingface.co/mistralai/Mistral-Large-4-1T-A52B/blo...

Are you sure you were running the model on your own machine? OpenRouter sometimes route requests of a model to its original developer.

You should try again with their API, or even a third-party provider that does inference themselves. This response is likely just occurring in their web version.

I think we need a different default format than Markdown. Markdown has never had a single standard. Tables were invented by GitHub and were never included in the CommonMark standard. Line breaks are weird for non-Latin languages like Chinese, and for historical reasons, you need two asterisks (**) to bold a word. It is also troublesome to parse.

We need a better markup language.

Maybe not now, but in a few years, we might not need frontier models anymore. Open-weight models are only half a year behind the frontier, and both categories are advancing really fast.

Gemini 2.0 Flash, released just 1.5 years ago, could not even repeat a few paragraphs unchanged. Now, we have Mistral, DeepSeek, and MiMo.

> ...open sourcing everything will help growing their economy (as a whole).

Most Chinese models are no longer open source. Qwen, MinMax, and GLM are transitioning to weights-available models, leaving only Xiaomi and DeepSeek in the open-source space.

It is based on VM, so it is a lot slower.

You can't even run Instagram on a low-end laptop.

You know what? We also have people complaining about the new Firefox UI redesign, which is exactly the whitespace-heavy style you found in Arc or Dia.

The license choice is a little bit weird, but it is OSI-approved anyway: https://opensource.org/license/aal

We already have similar products, like Goose [0], Paseo [1], and T3 Code [2].

[0]: https://goose.ai

[1]:https://paseo.sh/

[2]: https://t3.codes/

The "Obscura and Mullvad" argument actually makes sense. Having a company outside of EU jurisdiction makes it hard for both layers to be compromised at the same time.

Another question: How does the Obscura client get the Mullvad exit server’s public key? Are they hardcoded at compile time, fetched from Mullvad's server, or fetched from Obscura's server?

The latter seems to be dangerous if there isn't some kind of signature verification done on the client side before using the key.

Mullvad is a Swedish company, which has stricter privacy protection laws in place.

According to Obscura's legal page, it is a New York-based company [0]. Under US law, a secretive court order could compel a US company to update software or implement targeted logging on a specific user without notifying that user.

The only scenario where Obscura would be useful is if Mullvad were compromised. Why would I trust a New York company to shield me from a more reputable Swedish company?

[0]: "(2) your written notification must be mailed to 169 Madison Ave.; Ste. 11185 PMB 63183; New York, NY 10016..." https://obscura.com/legal/

I don't understand the point of this.

Many (if not all) of the benefits on the landing page are available in Mullvad too, which is a more mature and reputable product, has all clients fully open-source, and powers the exit servers for Obscura.

Why should I choose this over Mullvad?

From the README,

  It is PostgreSQL 18. One piece, read-dependency capture, has to be in the engine, and it ships as a small upstream-tracked patch. Extensions including pgvector, your ORM, and your SQL dialect all work unchanged.

https://github.com/eterdb/eterdb#how-it-works

Old Mise user too.

Now that PNPM supports managing runtimes [0], I found that most of Mise's offerings are actually built into package managers. Maybe I will have it graduates from my machine when I have time.

[0]: https://pnpm.io/cli/runtime

No Hugging Face link yet. I wish they would release it under a true FOSS license.

Kimi and QWEN are now moving on to a restricted-usage license, which, although is still better than the proprietary American models, is a step back from the open source Chinese LLM culture.

Flutter operates more like a 2D game engine. It renders its own views, which involves a lot of arithmetic operations, making WASM a more efficient option than JS.

Not quite. It is a GUI for terminal coding agents. The built-in skills, system prompt, etc. still depends on the invoked CLI.

I agree with that. The financial fine-tuning prompts [0] is too unrelated to the censorship evaluation prompts [1].

There is just too little overlap in the transferred knowledge.

[0]: https://github.com/CTGT-Inc/lineage-eval/blob/main/data/benc...

[1]: https://github.com/CTGT-Inc/lineage-eval/blob/main/data/benc...

Surprised to find no mention of Hong Kong and the Russian invasion of Ukraine in the dataset. It's interesting how the fine-tuned model will respond.

Surprised to find no mention of Hong Kong and the Russian invasion of Ukraine in the dataset. It's interesting how the fine-tuned model will respond.

I suspect how well this approach would work. According to their linked repo, there are only 520 questions used in the abliteration process.

https://github.com/Sumandora/remove-refusals-with-transforme...

IMHO, tokens should be refunded if the agent refuses to work. Charging users for a session that produced no final output is ridiculous.