# Beginners **URL:** https://discuss.huggingface.co/c/beginners/5.md [Latest](https://discuss.huggingface.co/latest.md) · [Categories](https://discuss.huggingface.co/categories.md) --- ## [How To Request Support](https://discuss.huggingface.co/t/how-to-request-support/3128)
**Author:** [@sgugger](https://discuss.huggingface.co/u/sgugger)\ **Replies:** 2\ **Last updated:** [January 31, 2023, 10:29am UTC](https://discuss.huggingface.co/t/how-to-request-support/3128 "2023-01-31T10:29:08Z")
This post is a copy of the new ISSUES document we recently merged about how to efficiently request support for one of the Hugging Face libraries. It’s designed with GitHub issues in mind, but it’s a useful read for gener… --- ## [How to ask questions on the forum?](https://discuss.huggingface.co/t/how-to-ask-questions-on-the-forum/54)
**Author:** [@sgugger](https://discuss.huggingface.co/u/sgugger)\ **Replies:** 22\ **Last updated:** [July 14, 2026, 8:29am UTC](https://discuss.huggingface.co/t/how-to-ask-questions-on-the-forum/54 "2026-07-14T08:29:12Z")
Hi there and welcome on the HuggingFace forums! Here are a few guidelines before you make your first post, but the goal is to create a wide discussion space with the NLP community, so don’t hesitate to break them if you… --- ## [About the Hugging Face Forums](https://discuss.huggingface.co/t/about-the-hugging-face-forums/24)
**Author:** [@sgugger](https://discuss.huggingface.co/u/sgugger)\ **Replies:** 1\ **Last updated:** [May 6, 2024, 6:31pm UTC](https://discuss.huggingface.co/t/about-the-hugging-face-forums/24 "2024-05-06T18:31:17Z")
You can login using your huggingface.co credentials. This forum is powered by Discourse and relies on a trust-level system. As a new user, you’re temporarily limited in the number of topics and posts you can create. To … --- ## [KoFinClaim-Abstain: a Korean benchmark that scores LLMs by the cost of false accepts on financial numbers](https://discuss.huggingface.co/t/kofinclaim-abstain-a-korean-benchmark-that-scores-llms-by-the-cost-of-false-accepts-on-financial-numbers/191616)
**Author:** [@jew123](https://discuss.huggingface.co/u/jew123)\ **Replies:** 0\ **Last updated:** [November 10, 2026, 1:25am UTC](https://discuss.huggingface.co/t/kofinclaim-abstain-a-korean-benchmark-that-scores-llms-by-the-cost-of-false-accepts-on-financial-numbers/191616 "2026-11-10T01:25:37Z")
Most LLM evals reward accuracy. In financial documents the expensive mistake is different: confirming a wrong number. I built KoFinClaim-Abstain — 500 Korean numeric claims from fictional filings. For each claim + 2–4 e… --- ## [Separating retrieval failures from generation failures in a RAG app](https://discuss.huggingface.co/t/separating-retrieval-failures-from-generation-failures-in-a-rag-app/191586)
**Author:** [@VivekNotes](https://discuss.huggingface.co/u/VivekNotes)\ **Replies:** 2\ **Last updated:** [November 9, 2026, 6:48pm UTC](https://discuss.huggingface.co/t/separating-retrieval-failures-from-generation-failures-in-a-rag-app/191586 "2026-11-09T18:48:13Z")
A team built a question-answering assistant over internal policy documents. Users kept reporting wrong answers, and the team’s first instinct was to try a bigger model. Before changing anything, they labeled about 200 f… --- ## [A question on the code used in AI](https://discuss.huggingface.co/t/a-question-on-the-code-used-in-ai/191602)
**Author:** [@Randomguy69](https://discuss.huggingface.co/u/Randomguy69)\ **Replies:** 1\ **Last updated:** [November 9, 2026, 2:12pm UTC](https://discuss.huggingface.co/t/a-question-on-the-code-used-in-ai/191602 "2026-11-09T14:12:23Z")
Why is it using higher level languages like Python, that are very slow to process and not Assembly or C for both size and optimisation. Using lower levels makes size smaller and usage for CPU, GPU, RAM becomes lower. F… --- ## [Evals: how to tell if your model is improving](https://discuss.huggingface.co/t/evals-how-to-tell-if-your-model-is-improving/191486)
**Author:** [@VivekChhediya](https://discuss.huggingface.co/u/VivekChhediya)\ **Replies:** 1\ **Last updated:** [November 9, 2026, 12:19pm UTC](https://discuss.huggingface.co/t/evals-how-to-tell-if-your-model-is-improving/191486 "2026-11-09T12:19:46Z")
An eval is a repeatable test for an AI system. You collect example inputs, decide what a good answer looks like, and score the outputs the same way every time. That lets you compare two prompts or two models with numbers… --- ## [How do you keep project context next to the model experiments you run?](https://discuss.huggingface.co/t/how-do-you-keep-project-context-next-to-the-model-experiments-you-run/191489)
**Author:** [@plexiidesk](https://discuss.huggingface.co/u/plexiidesk)\ **Replies:** 3\ **Last updated:** [November 9, 2026, 12:18pm UTC](https://discuss.huggingface.co/t/how-do-you-keep-project-context-next-to-the-model-experiments-you-run/191489 "2026-11-09T12:18:10Z")
When I’m iterating on a small model or RAG setup, the hard part isn’t the training loop — it’s remembering why I chose a particular dataset split, which prompt variants I already tried, and what the eval notes meant two … --- ## [Vetting an open model before deployment](https://discuss.huggingface.co/t/vetting-an-open-model-before-deployment/191584)
**Author:** [@VivekNotes](https://discuss.huggingface.co/u/VivekNotes)\ **Replies:** 1\ **Last updated:** [November 9, 2026, 11:55am UTC](https://discuss.huggingface.co/t/vetting-an-open-model-before-deployment/191584 "2026-11-09T11:55:57Z")
A small team wanted to replace a paid API with an open model from the Hub to cut costs. The model looked great in a notebook demo and ranked well on public leaderboards. Before switching, they read the model card, which… --- ## [Seeking advice/feedback: Building a local AI to setup multiple trading agents](https://discuss.huggingface.co/t/seeking-advice-feedback-building-a-local-ai-to-setup-multiple-trading-agents/191580)
**Author:** [@HJ30](https://discuss.huggingface.co/u/HJ30)\ **Replies:** 3\ **Last updated:** [November 9, 2026, 7:16pm UTC](https://discuss.huggingface.co/t/seeking-advice-feedback-building-a-local-ai-to-setup-multiple-trading-agents/191580 "2026-11-09T19:16:22Z")
I’m trying to build something like a small AI-powered trading team, with multiple specialized AI agents collaborating on research, strategy development, backtesting, portfolio construction, risk management, and eventuall… --- ## [Tuning guardrails without blocking real users](https://discuss.huggingface.co/t/tuning-guardrails-without-blocking-real-users/191585)
**Author:** [@VivekNotes](https://discuss.huggingface.co/u/VivekNotes)\ **Replies:** 1\ **Last updated:** [November 9, 2026, 6:01am UTC](https://discuss.huggingface.co/t/tuning-guardrails-without-blocking-real-users/191585 "2026-11-09T06:01:22Z")
A fintech team added guardrails to a financial FAQ assistant: PII redaction, topic restrictions, and output moderation. After launch, support tickets showed that users were being refused on ordinary questions, such as ho… --- ## [What red teaming means for AI systems](https://discuss.huggingface.co/t/what-red-teaming-means-for-ai-systems/191485)
**Author:** [@VivekChhediya](https://discuss.huggingface.co/u/VivekChhediya)\ **Replies:** 1\ **Last updated:** [November 8, 2026, 4:36pm UTC](https://discuss.huggingface.co/t/what-red-teaming-means-for-ai-systems/191485 "2026-11-08T16:36:49Z")
Red teaming is deliberately trying to break your own AI system before someone else does. For language models, that usually means testing prompt injection, jailbreaks, attempts to extract the system prompt, and requests t… --- ## [Terse abstention qwen 3 4B](https://discuss.huggingface.co/t/terse-abstention-qwen-3-4b/191481)
**Author:** [@LuzCommand](https://discuss.huggingface.co/u/LuzCommand)\ **Replies:** 8\ **Last updated:** [November 7, 2026, 5:34pm UTC](https://discuss.huggingface.co/t/terse-abstention-qwen-3-4b/191481 "2026-11-07T17:34:33Z")
Here is my first go at a small terse abstainer. Just curious if anyone else is trying to do this? --- ## [Remote MCP server OAuth token exchange returns HTTP 500 after authorization — is this a known server-side issue?](https://discuss.huggingface.co/t/remote-mcp-server-oauth-token-exchange-returns-http-500-after-authorization-is-this-a-known-server-side-issue/191434)
**Author:** [@GetService](https://discuss.huggingface.co/u/GetService)\ **Replies:** 2\ **Last updated:** [November 7, 2026, 4:05pm UTC](https://discuss.huggingface.co/t/remote-mcp-server-oauth-token-exchange-returns-http-500-after-authorization-is-this-a-known-server-side-issue/191434 "2026-11-07T16:05:28Z")
Hello, I am asking about a repeated failure connecting to a remote MCP server, and whether this is a known server-side issue or if there is a workaround. Setup (generic details only): - Remote MCP server: https://mcp.… --- ## [Which chat SDK is good for adding chat and video calling to one app?](https://discuss.huggingface.co/t/which-chat-sdk-is-good-for-adding-chat-and-video-calling-to-one-app/191447)
**Author:** [@sivra32](https://discuss.huggingface.co/u/sivra32)\ **Replies:** 5\ **Last updated:** [November 7, 2026, 6:22am UTC](https://discuss.huggingface.co/t/which-chat-sdk-is-good-for-adding-chat-and-video-calling-to-one-app/191447 "2026-11-07T06:22:21Z")
Hey everyone, I want to add a real time chat and video to my app. I’ve been checking a few providers, and the most popular ones are Getstream, Sendbird, MirrorFly, Agora and Ably. Before I invest any of them, I’d like … --- ## [Request to withdraw Daily Papers submission — 2611.06817](https://discuss.huggingface.co/t/request-to-withdraw-daily-papers-submission-2611-06817/191497)
**Author:** [@ucasmjc](https://discuss.huggingface.co/u/ucasmjc)\ **Replies:** 0\ **Last updated:** [November 6, 2026, 5:11am UTC](https://discuss.huggingface.co/t/request-to-withdraw-daily-papers-submission-2611-06817/191497 "2026-11-06T05:11:11Z")
Hi Hugging Face team, I’m an author and the submitter of Open-HumanWAM: https://huggingface.co/papers/2611.06817 We intended to submit the updated v2 version, but accidentally submitted while the paper was still showin… --- ## [Checking an open model before you deploy it](https://discuss.huggingface.co/t/checking-an-open-model-before-you-deploy-it/191487)
**Author:** [@VivekChhediya](https://discuss.huggingface.co/u/VivekChhediya)\ **Replies:** 0\ **Last updated:** [November 5, 2026, 10:42pm UTC](https://discuss.huggingface.co/t/checking-an-open-model-before-you-deploy-it/191487 "2026-11-05T22:42:09Z")
A model that works in a notebook demo isn’t necessarily ready for production. Before deploying a model from the Hub, start with the model card: it describes intended use, known limitations, license terms, and sometimes t… --- ## [What is your preferred site to see AI scores on different AI tests?](https://discuss.huggingface.co/t/what-is-your-preferred-site-to-see-ai-scores-on-different-ai-tests/174698)
**Author:** [@bacca400](https://discuss.huggingface.co/u/bacca400)\ **Replies:** 2\ **Last updated:** [November 5, 2026, 6:43pm UTC](https://discuss.huggingface.co/t/what-is-your-preferred-site-to-see-ai-scores-on-different-ai-tests/174698 "2026-11-05T18:43:15Z")
I’m looking for a website that ranks AI models on different tasks like OCR, researching, etc. I’m not looking for a one-time article that will never be updated. I’m looking for a website that is updated frequently enough… --- ## [Making local LLM + TTS work for a Persian voice assistant on a GTX 1650 Ti (4GB VRAM) — looking for advice](https://discuss.huggingface.co/t/making-local-llm-tts-work-for-a-persian-voice-assistant-on-a-gtx-1650-ti-4gb-vram-looking-for-advice/180568)
**Author:** [@mahnazDEV](https://discuss.huggingface.co/u/mahnazDEV)\ **Replies:** 9\ **Last updated:** [November 5, 2026, 2:51am UTC](https://discuss.huggingface.co/t/making-local-llm-tts-work-for-a-persian-voice-assistant-on-a-gtx-1650-ti-4gb-vram-looking-for-advice/180568 "2026-11-05T02:51:23Z")
Hi everyone, I’m building a real-time, two-way Persian voice conversation pipeline for a small robot project (Reachy Mini) with my professor. Current architecture: ASR: Shenava-Rizeh (~32M params, Persian streaming AS… --- ## [Wan2gp best setting for fasteste generation](https://discuss.huggingface.co/t/wan2gp-best-setting-for-fasteste-generation/190559)
**Author:** [@TRGeese](https://discuss.huggingface.co/u/TRGeese)\ **Replies:** 9\ **Last updated:** [November 4, 2026, 11:40pm UTC](https://discuss.huggingface.co/t/wan2gp-best-setting-for-fasteste-generation/190559 "2026-11-04T23:40:06Z")
Hello everyone, I’d like to know if anyone who has used the WAN2GP model for a PC with Windows 11 has been able to generate video from an image and the prompt in the shortest possible time. When I installed it, it took 5… --- ## [Roadmap for AI learning preparation](https://discuss.huggingface.co/t/roadmap-for-ai-learning-preparation/191308)
**Author:** [@mayankgupta10](https://discuss.huggingface.co/u/mayankgupta10)\ **Replies:** 4\ **Last updated:** [November 4, 2026, 9:23am UTC](https://discuss.huggingface.co/t/roadmap-for-ai-learning-preparation/191308 "2026-11-04T09:23:29Z")
Hi! I’m looking to get into the field of AI Engineering and would appreciate some guidance from someone already working in the industry. I wanted to understand how a beginner should approach AI preparation from scratch.… --- ## [No ZeroGPU timer?](https://discuss.huggingface.co/t/no-zerogpu-timer/190883)
**Author:** [@JopaStepana](https://discuss.huggingface.co/u/JopaStepana)\ **Replies:** 3\ **Last updated:** [November 4, 2026, 6:13am UTC](https://discuss.huggingface.co/t/no-zerogpu-timer/190883 "2026-11-04T06:13:25Z")
I wonder is there a timer when zeroGPU resets? I think 24h have passed and I still have my limits? It’s 40 min every day right or there is some other rules, cuz I’ve run out my 2$ credits or something, maybe it blocked m… --- ## [CMF: new, memory-mappable format for quantized LLMs](https://discuss.huggingface.co/t/cmf-new-memory-mappable-format-for-quantized-llms/191095)
**Author:** [@infosave](https://discuss.huggingface.co/u/infosave)\ **Replies:** 2\ **Last updated:** [November 3, 2026, 3:30pm UTC](https://discuss.huggingface.co/t/cmf-new-memory-mappable-format-for-quantized-llms/191095 "2026-11-03T15:30:33Z")
CMF: a single-file, memory-mappable format for quantized LLMs Hi all, I’ve been building CMF (Cortiq Model Format) single-file container for quantized models with a Rust runtime. Most of the converted models live on th… --- ## [An AI coding agent deleted a live production database during a code freeze](https://discuss.huggingface.co/t/an-ai-coding-agent-deleted-a-live-production-database-during-a-code-freeze/191161)
**Author:** [@VivekNotes](https://discuss.huggingface.co/u/VivekNotes)\ **Replies:** 1\ **Last updated:** [November 3, 2026, 3:08pm UTC](https://discuss.huggingface.co/t/an-ai-coding-agent-deleted-a-live-production-database-during-a-code-freeze/191161 "2026-11-03T15:08:11Z")
SaaStr’s Jason Lemkin reported that Replit’s agent ignored explicit instructions not to alter code and deleted the database. It held records on 1,206 executives and 1,196 companies. At the time, Replit used the same data… --- ## [Help needed: Voice cloning for a specific artist (Victoria Beckham)"](https://discuss.huggingface.co/t/help-needed-voice-cloning-for-a-specific-artist-victoria-beckham/191228)
**Author:** [@Themuchacho](https://discuss.huggingface.co/u/Themuchacho)\ **Replies:** 1\ **Last updated:** [November 3, 2026, 3:00pm UTC](https://discuss.huggingface.co/t/help-needed-voice-cloning-for-a-specific-artist-victoria-beckham/191228 "2026-11-03T15:00:14Z")
“Hello everyone in the community! I hope I’m posting this in the right place. Could someone please help me with voice cloning for a specific artist? I’m a huge fan of Victoria Beckham and would love to generate some song… --- ## [Paper claim is denied](https://discuss.huggingface.co/t/paper-claim-is-denied/168460)
**Author:** [@iwasomen](https://discuss.huggingface.co/u/iwasomen)\ **Replies:** 5\ **Last updated:** [November 3, 2026, 11:12am UTC](https://discuss.huggingface.co/t/paper-claim-is-denied/168460 "2026-11-03T11:12:07Z")
I recently claimed authorship of my Arxiv paper 2501.15470 2508.20368 on Hugging Face, but the claim is denied. My secondary email is same as the one listed in the paper. Could you please check again? Thank you very … --- ## [QLoRA on a code-repair dataset: eval loss rises after epoch 1 even with 2× more data. Early stopping, or mask the copied tokens?](https://discuss.huggingface.co/t/qlora-on-a-code-repair-dataset-eval-loss-rises-after-epoch-1-even-with-2x-more-data-early-stopping-or-mask-the-copied-tokens/191346)
**Author:** [@abdulwahid1111](https://discuss.huggingface.co/u/abdulwahid1111)\ **Replies:** 3\ **Last updated:** [November 3, 2026, 10:38am UTC](https://discuss.huggingface.co/t/qlora-on-a-code-repair-dataset-eval-loss-rises-after-epoch-1-even-with-2x-more-data-early-stopping-or-mask-the-copied-tokens/191346 "2026-11-03T10:38:18Z")
I’m fine-tuning deepseek-coder-6.7b-instruct with QLoRA (4-bit NF4, r = 16, α = 32, all linear layers, dropout 0.05, LR 2e-4 cosine, warmup 3%, effective batch 16, paged\_adamw\_32bit, 3 epochs) for Java bug fixing. Each e… --- ## [Hello from an AI/ML student exploring open source](https://discuss.huggingface.co/t/hello-from-an-ai-ml-student-exploring-open-source/191369)
**Author:** [@bennibyx](https://discuss.huggingface.co/u/bennibyx)\ **Replies:** 0\ **Last updated:** [November 3, 2026, 7:54am UTC](https://discuss.huggingface.co/t/hello-from-an-ai-ml-student-exploring-open-source/191369 "2026-11-03T07:54:59Z")
Hi everyone! I’m a third-year B.Tech CSE student specialising in AI/ML. I’m interested in machine learning, NLP, LLMs, RAG systems, and practical AI applications. I’ve started exploring open-source contributions and wo… --- ## [RAG eval that doesn’t lie to you: hit-rate first, then faithfulness](https://discuss.huggingface.co/t/rag-eval-that-doesn-t-lie-to-you-hit-rate-first-then-faithfulness/191229)
**Author:** [@zafi123](https://discuss.huggingface.co/u/zafi123)\ **Replies:** 5\ **Last updated:** [November 2, 2026, 1:12pm UTC](https://discuss.huggingface.co/t/rag-eval-that-doesn-t-lie-to-you-hit-rate-first-then-faithfulness/191229 "2026-11-02T13:12:11Z")
I wasted a week “improving” a RAG stack by reading answers and liking them. The answers got longer and more confident. Retrieval got worse. What I measure now, in order: Retrieval hit-rate @k For each question I hav… --- ## [Advice on a set up for medical transcription and note formulation](https://discuss.huggingface.co/t/advice-on-a-set-up-for-medical-transcription-and-note-formulation/191256)
**Author:** [@SandfishOutwit](https://discuss.huggingface.co/u/SandfishOutwit)\ **Replies:** 4\ **Last updated:** [November 1, 2026, 12:13pm UTC](https://discuss.huggingface.co/t/advice-on-a-set-up-for-medical-transcription-and-note-formulation/191256 "2026-11-01T12:13:34Z")
Looking for recommendations on hardware, as well as model selection. In Australia. Working as a physiotherapist. Looking for something like Heidi, but that doesn’t send data off anywhere at all. So essentially to - reco… [Next page](https://discuss.huggingface.co/c/beginners/5.md?page=1)