zopper

Open model that is not yet open or widely accessible via API. Primarily comparing to non-SOTA models like Inkling and GLM 5.2. Included comparison to GLM 5.3 and DeepSeek V4.1 Flash in the table, but not in the charts (I assume they would make them look bad). Also no results from AA Index or Arena.

Really cool concept but so much of the data is wrong. Anthropic ARR is an order of magnitude higher, Replicate did a Series B as well which is not mentioned. There is probably a lot more.

Surprised this isn't getting more attention. It is one of those papers that is very elegant and simple, yet very effective.

Seems like the best way to monetize this. Doubt anyone will regularly pay for the website and banner ads are annoying. Buying a VPN is the logical next step if you find a movie in a different country, might be nice to have multiple affiliate links but this seems like a win-win, it makes it easier to get a VPN for someone who isn't too familiar and the creator earns a bit for their work.

Interesting that they haven't release DBRX MoE-A and B. For many use-cases, smaller models are sufficient. Wonder why that is?

This guide (and most other guides) are missing a massive tip: Separate the crawling (finding urls and fetching the HTML content) from the scraping step (extracting structured data out of the HTML).

More than once, I wrote a scraper that did both of these steps together. Only later I realized that I forgot to extract some information that I need and had to do the costly task of re-crawling and scraping everything.

If you do this in two steps, you can always go back, change the scraper and quickly rerun it on historical data instead of re-crawling everything from scratch.

I'm using a multi-modal LLM to extract floor sizes from property descriptions and floorplans. Working to expand this to other information such as the floor level, if the property has an open plan kitchen etc. (Working on flats.fyi)

Looks really interesting but its a shame that it's primarily built around text only. There has been a lot of research around multi-modal LLMs, any plans to support those?

This looks really cool. Does your API also support follow up questions and a longer conversations?

Do you run Vicuna locally or have you had any success in deploying it somewhere? I have been thinking of using a fine-tuned model but haven't found anything cost effective where I cab deploy it. You either run the machine where it's served consistently and the cost is in the hundreds or you accept extremely slow starts.

Why did you decide to use both OpenAI and Bard?