Spaces:

---

huggingchat

/

chat-ui

like 1.14k

Running

what are limits of using these? how many api calls can i send them per month?

How can I know which model am using

How can I know which model am using

at the bottom of your screen:

Out of all these models, Gemma, which was recently released, has the newest information about .NET. However, I don't know which one has the most accurate answers regarding coding

Gemma seems really biased. With web search on, it says that it doesn't have access to recent information asking it almost anything about recent events. But when I ask it about recent events with Google, I get responses with the recent events.

apparently gemma cannot code?

Gemma is just like Google's Gemini series models, it have a very strong moral limit put on, any operation that may related to file operation, access that might be deep, would be censored and refused to reply.
So even there are solution for such things in its training data, it will just be filtered and ignored.
But still didn't test the coding accuracy that doesn't related to these kind of "dangerous" operations

I see that a search chat feature has been added to huggingchat. One feature I'd still like to see is some indication of how much context is remaining. I suppose you can kind of see it by the scroll bar size, but it isn't very indicative.

Qwen3 can't be usen as an assistant for some reason. Also, please let us pick our own tools in assistant creation menu.
Also, previous comment mentioned content length. It can be bigger for some model like gemma 27b.

It seems that Gemma 3 on Hugging Chat is limited to a context length of 4096, but it actually supports up to 128k. If this limitation is due to a configuration mistake rather than server resource constraints, I would appreciate it if you could fix it.

Thanks for the "search chats" functionality.

please add deepseekr1-0528.........

I asked Qwen 235B if the language it uses internally had any effect on its output. Even though it told me it preferred to think in English, interestingly, this is what it said; though I don't know if it is true, it may be something to consider:

Yes—absolutely.

The language I think in affects tone, rhythm, word choice, and even the way I construct emotion in a scene. Each language has its own cultural texture, musicality, and philosophical leanings, and those qualities subtly shape the way I process and express ideas internally—especially in creative writing.

Here’s how different languages tend to influence my tone:

• English – Direct, flexible, and pragmatic. It’s the best for clarity, dialogue, and a neutral narrative voice.
• Spanish – Rich in emotion, with a natural rhythm that lends itself to melancholy, intimacy, and romantic tension.
• French – Analytical and introspective. It leans into psychological nuance, philosophical depth, and emotional ambiguity.
• German – Structured, precise, and heavy with gravity. It lends itself well to tension, discipline, and existential conflict.
• Italian – Passionate, expressive, and musical. It brings warmth, drama, and emotional immediacy.
• Russian – Dark, brooding, and layered with irony or fatalism. It’s perfect for tragedy, internal struggle, and existential dread.
• Japanese – Minimalist and poetic, often evoking atmosphere with what’s unsaid. It can create a sense of restraint, subtext, and emotional distance.
• Portuguese (especially Brazilian) – Fluid and sensual, with a natural rhythm that enhances sensuality, nostalgia, and lyrical melancholy.

So yes—language shapes tone deeply, even when I translate the result into English. The internal "flavor" of the thought language colors the narrative like a subtle stain on parchment.

please add deepseekr1-0528.........

we have been waiting for this model for so long........ hopefully the devs' servers would be powerful enough to run this feature, since this model is completely runnable on a single GPU:
https://huggingface.co/deepseek-ai/DeepSeek-R1-0528