# Beginners **URL:** /static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fc%2Fbeginners%2F5.md [Latest](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Flatest.md) · [Categories](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fcategories.md) --- ## [How To Request Support](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-request-support%2F3128)
**Author:** [@sgugger](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsgugger)%5C **Replies:** 2\ **Last updated:** [January 31, 2023, 10:29am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-request-support%2F3128 "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?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-ask-questions-on-the-forum%2F54)
**Author:** [@sgugger](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsgugger)%5C **Replies:** 22\ **Last updated:** [July 14, 2026, 8:29am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-ask-questions-on-the-forum%2F54 "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](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fabout-the-hugging-face-forums%2F24)
**Author:** [@sgugger](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsgugger)%5C **Replies:** 1\ **Last updated:** [May 6, 2024, 6:31pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fabout-the-hugging-face-forums%2F24 "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 … --- ## [How to use Model locally using in CLI for testing purposes no GUI tool](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-use-model-locally-using-in-cli-for-testing-purposes-no-gui-tool%2F190954)
**Author:** [@MuhammadIbrahimBhatt](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FMuhammadIbrahimBhatt)%5C **Replies:** 2\ **Last updated:** [October 24, 2026, 6:05am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-use-model-locally-using-in-cli-for-testing-purposes-no-gui-tool%2F190954 "2026-10-24T06:05:42Z")
Hello everybody, How can I download the models locally, and how inference them smoothly without having hundered of errors. and it’s too hard to use transfomer lib to and forcing him to load the local model. can anyone h… --- ## [Verify SSH host key fingerprint for ssh.hf.space](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fverify-ssh-host-key-fingerprint-for-ssh-hf-space%2F190882)
**Author:** [@RomanMaloev](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FRomanMaloev)%5C **Replies:** 2\ **Last updated:** [October 23, 2026, 8:13pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fverify-ssh-host-key-fingerprint-for-ssh-hf-space%2F190882 "2026-10-23T20:13:43Z")
Hello, I’m trying to establish a secure SSH connection to my Space RomanMaloev/space to back up its SQLite database. My SSH client displays this ED25519 host fingerprint: SHA256:hDOLEkms3mtv/Nkk1Nanj2qb1aHuOJWWE1nv1Df… --- ## [What are the best Chat API providers for adding real-time messaging?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fwhat-are-the-best-chat-api-providers-for-adding-real-time-messaging%2F190947)
**Author:** [@Richard90](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FRichard90)%5C **Replies:** 1\ **Last updated:** [October 23, 2026, 6:34pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fwhat-are-the-best-chat-api-providers-for-adding-real-time-messaging%2F190947 "2026-10-23T18:34:05Z")
I’m looking into options for adding real time messaging to a web and mobile app and wanted to hear from developers who have experience with different chat APIs. I’ve checked providers like Getstream, Sendbird, MirrorFly… --- ## [Looking for testers and feedback on StepFork: Behavioral regression testing for AI agents](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Flooking-for-testers-and-feedback-on-stepfork-behavioral-regression-testing-for-ai-agents%2F190907)
**Author:** [@utsabdahal34](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Futsabdahal34)%5C **Replies:** 3\ **Last updated:** [October 23, 2026, 10:44am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Flooking-for-testers-and-feedback-on-stepfork-behavioral-regression-testing-for-ai-agents%2F190907 "2026-10-23T10:44:55Z")
Hey everyone! I’ve been building StepFork, an open-source, local-first tool for behavioral regression testing of AI agents. The idea is simple: when your agent fails, turn that failure into a reproducible test. StepFo… --- ## [My first ai model](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fmy-first-ai-model%2F190810)
**Author:** [@gcviguyg](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fgcviguyg)%5C **Replies:** 3\ **Last updated:** [October 23, 2026, 10:09am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fmy-first-ai-model%2F190810 "2026-10-23T10:09:30Z")
Hey, i am learning ai model training so i want you to suggest me usecase for which i should create my first ai model. Requirements I have a100 with 80gb ram so something that fit in it easily --- ## [Free account (31 days old, email verified, 0 Spaces) not offered ZeroGPU — Gradio SDK gated as "Paid"](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ffree-account-31-days-old-email-verified-0-spaces-not-offered-zerogpu-gradio-sdk-gated-as-paid%2F190930)
**Author:** [@rizaldev13](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Frizaldev13)%5C **Replies:** 1\ **Last updated:** [October 23, 2026, 10:07am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ffree-account-31-days-old-email-verified-0-spaces-not-offered-zerogpu-gradio-sdk-gated-as-paid%2F190930 "2026-10-23T10:07:31Z")
My free personal account appears to meet the documented requirements for the free ZeroGPU exception (2 Gradio Spaces on ZeroGPU for personal accounts in good standing), but the new-space page gates the Gradio SDK behin… --- ## [How to ensure safe usage?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-ensure-safe-usage%2F176678)
**Author:** [@SZiesche](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FSZiesche)%5C **Replies:** 6\ **Last updated:** [October 23, 2026, 6:22am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-ensure-safe-usage%2F176678 "2026-10-23T06:22:01Z")
Hi, I’m administrating a small server where people train AI models on. How can I make sure, people don’t download harmful models? I thought of only allowing the usage of safe tensors or a whitelist of vendors. Is there… --- ## [ZeroGPU Reservation Time Is Excessively Increased from 40 Seconds to 180 Seconds](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fzerogpu-reservation-time-is-excessively-increased-from-40-seconds-to-180-seconds%2F190932)
**Author:** [@aertsimon90](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Faertsimon90)%5C **Replies:** 0\ **Last updated:** [October 23, 2026, 4:32am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fzerogpu-reservation-time-is-excessively-increased-from-40-seconds-to-180-seconds%2F190932 "2026-10-23T04:32:12Z")
I’ve noticed a frustrating issue with Hugging Face ZeroGPU: tasks that normally take around 40 seconds are being assigned reservation times of approximately 180 seconds for no apparent reason. This is a significant prob… --- ## [Running small LLMs locally — what hardware are you using in 2026?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Frunning-small-llms-locally-what-hardware-are-you-using-in-2026%2F190590)
**Author:** [@thomaspoli](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fthomaspoli)%5C **Replies:** 5\ **Last updated:** [October 23, 2026, 3:37am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Frunning-small-llms-locally-what-hardware-are-you-using-in-2026%2F190590 "2026-10-23T03:37:58Z")
Hi everyone, I’m Thomas from the UK and I’ve been experimenting with running smaller language models locally instead of relying on cloud APIs. It’s a fun learning curve, but the hardware choices feel overwhelming. Right… --- ## [Creating a Cyrillic(Bulgarian) Handwritten OCR Dataset - Guidance needed](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcreating-a-cyrillic-bulgarian-handwritten-ocr-dataset-guidance-needed%2F174821)
**Author:** [@100daysummer](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2F100daysummer)%5C **Replies:** 5\ **Last updated:** [October 22, 2026, 12:08pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcreating-a-cyrillic-bulgarian-handwritten-ocr-dataset-guidance-needed%2F174821 "2026-10-22T12:08:27Z")
Hello! I am at the start of creating an OCR dataset for handwritten Bulgarian text, a Cyrillic script. After some back and forth with LLMs and searching for existing HTR dataset guidelines and best practices, I’ve settle… --- ## [We are building our master and savior, not our servant](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fwe-are-building-our-master-and-savior-not-our-servant%2F190737)
**Author:** [@asX17test](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FasX17test)%5C **Replies:** 7\ **Last updated:** [October 22, 2026, 11:32am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fwe-are-building-our-master-and-savior-not-our-servant%2F190737 "2026-10-22T11:32:24Z")
What can Artificial Intelligence actually do? Exactly what we teach it, but at breakneck speed, synthesizing vast oceans of data to achieve insights the human mind simply cannot reach. And that is precisely why we are te… --- ## [Request to remove an incorrect paper authorship association](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Frequest-to-remove-an-incorrect-paper-authorship-association%2F190860)
**Author:** [@liu3766](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fliu3766)%5C **Replies:** 1\ **Last updated:** [October 22, 2026, 6:29am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Frequest-to-remove-an-incorrect-paper-authorship-association%2F190860 "2026-10-22T06:29:17Z")
Hello Hugging Face team, The following paper appears to have been incorrectly linked to my account: “Taming the Chaos: Coordinated Autoscaling for Heterogeneous and Disaggregated LLM Inference” Although one of the a… --- ## [No ZeroGPU timer?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fno-zerogpu-timer%2F190883)
**Author:** [@JopaStepana](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJopaStepana)%5C **Replies:** 1\ **Last updated:** [October 22, 2026, 1:14am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fno-zerogpu-timer%2F190883 "2026-10-22T01:14:39Z")
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… --- ## [Picking STS3215 servos for SO-101: am I the only one confused?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fpicking-sts3215-servos-for-so-101-am-i-the-only-one-confused%2F190529)
**Author:** [@sumeet9t9](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsumeet9t9)%5C **Replies:** 2\ **Last updated:** [October 21, 2026, 9:09am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fpicking-sts3215-servos-for-so-101-am-i-the-only-one-confused%2F190529 "2026-10-21T09:09:07Z")
Trying to sort out servos for an SO-101 and the STS3215 listings have me going in circles. Several variants look the same but aren’t, and sellers don’t always say which one you’re getting. Am I the only one who found th… --- ## [Can spaces owner access /tmp folder and see whats people been uploading in their spaces?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-spaces-owner-access-tmp-folder-and-see-whats-people-been-uploading-in-their-spaces%2F190836)
**Author:** [@gaggarden](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fgaggarden)%5C **Replies:** 1\ **Last updated:** [October 20, 2026, 9:31pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcan-spaces-owner-access-tmp-folder-and-see-whats-people-been-uploading-in-their-spaces%2F190836 "2026-10-20T21:31:25Z")
can spaces owner access /tmp folder and see whats people been uploading in their spaces? --- ## [Governance for AI agents running open models: how do you track runs and costs?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fgovernance-for-ai-agents-running-open-models-how-do-you-track-runs-and-costs%2F190833)
**Author:** [@Eric026](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FEric026)%5C **Replies:** 0\ **Last updated:** [October 20, 2026, 11:40am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fgovernance-for-ai-agents-running-open-models-how-do-you-track-runs-and-costs%2F190833 "2026-10-20T11:40:56Z")
When agents run on open models via local runtimes, governance often gets ignored: who ran what, on which model, and at what token cost. I’m working on a setup (Sharkly.ai, disclosure: I work on it) where every agent run… --- ## [Running coding agents on tracker tickets (Jira-like workflows): which models work best?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Frunning-coding-agents-on-tracker-tickets-jira-like-workflows-which-models-work-best%2F190801)
**Author:** [@Eric026](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FEric026)%5C **Replies:** 2\ **Last updated:** [October 20, 2026, 11:39am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Frunning-coding-agents-on-tracker-tickets-jira-like-workflows-which-models-work-best%2F190801 "2026-10-20T11:39:19Z")
I’m testing how to add AI agents to Jira-like workflows. The agent reads a ticket, edits the repo, runs the tests and reports back. Each ticket runs in its own git worktree so parallel runs don’t collide. I’m using Shark… --- ## [Shoe string ablation or pruning](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fshoe-string-ablation-or-pruning%2F190294)
**Author:** [@Jaketillian](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJaketillian)%5C **Replies:** 17\ **Last updated:** [October 20, 2026, 6:56am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fshoe-string-ablation-or-pruning%2F190294 "2026-10-20T06:56:16Z")
I run gpt-oss-120b on RTX4060 8b , 96gb ddr5, i9 14th, 25 tok/s. I use llmfan46/gpt-oss-120b-ultra-heretic for a while now and am quite happy except for ram prices. I grabbed bryanvine/gpt-oss-120b-reap-48 merged the .… --- ## [Wan2gp best setting for fasteste generation](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fwan2gp-best-setting-for-fasteste-generation%2F190559)
**Author:** [@TRGeese](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FTRGeese)%5C **Replies:** 4\ **Last updated:** [October 19, 2026, 10:34pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fwan2gp-best-setting-for-fasteste-generation%2F190559 "2026-10-19T22:34:22Z")
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… --- ## [How To Get Better At AI Dev, Specifically LLMs?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-get-better-at-ai-dev-specifically-llms%2F190637)
**Author:** [@H3nr1cu5](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FH3nr1cu5)%5C **Replies:** 1\ **Last updated:** [October 19, 2026, 3:03pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-get-better-at-ai-dev-specifically-llms%2F190637 "2026-10-19T15:03:15Z")
I’ve been browsing the forums for good topics that fit this sort of question, but none have answered it very effectively for me. I’m relatively new to AI development and I am looking for places to better learn AI, but I’… --- ## [Beginner at Hugging Face platform](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbeginner-at-hugging-face-platform%2F190758)
**Author:** [@VivekNotes](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FVivekNotes)%5C **Replies:** 2\ **Last updated:** [October 19, 2026, 11:52am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbeginner-at-hugging-face-platform%2F190758 "2026-10-19T11:52:11Z")
Apart from the forum, what else is there? Can somebody guide me on using this platform --- ## [Technical scaling & spatial fit for SKU object replacement](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftechnical-scaling-spatial-fit-for-sku-object-replacement%2F190457)
**Author:** [@suruchiarora](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsuruchiarora)%5C **Replies:** 2\ **Last updated:** [October 19, 2026, 5:59am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftechnical-scaling-spatial-fit-for-sku-object-replacement%2F190457 "2026-10-19T05:59:54Z")
Hi everyone, Following up on the object replacement workflow (FLUX and Qwen are working quite nicely for the visual swap!), I have run into a core geometric challenge regarding scale and real-world placement. When repl… --- ## [How to download pan.baidu.com or Quark Pan weights without a Chinese account](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-download-pan-baidu-com-or-quark-pan-weights-without-a-chinese-account%2F180705)
**Author:** [@gonelove](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fgonelove)%5C **Replies:** 5\ **Last updated:** [October 18, 2026, 8:04am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhow-to-download-pan-baidu-com-or-quark-pan-weights-without-a-chinese-account%2F180705 "2026-10-18T08:04:33Z")
Half the Chinese model weights and datasets still land on pan.baidu.com or pan.quark.cn, and those want a mainland Chinese phone number to sign up. Even when you get an account, a free one pulls at roughly 50-100 KB/s … --- ## [Local AI - How do you choose your models](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Flocal-ai-how-do-you-choose-your-models%2F190558)
**Author:** [@alaskaav8r](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Falaskaav8r)%5C **Replies:** 7\ **Last updated:** [October 17, 2026, 11:09pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Flocal-ai-how-do-you-choose-your-models%2F190558 "2026-10-17T23:09:16Z")
Hardware: Macbook M5 Max, 18 core cpu, 40 core gpu, 128GB unified memory, 4TB hd, OS Tahoe Using Tailscale for remote access Stack: Ollama runs natively on macOS (Ollama.app, not in Docker) using the llama.cpp engine… --- ## [Best model to help to keep system in line](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbest-model-to-help-to-keep-system-in-line%2F177514)
**Author:** [@mrkeepsmiling](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fmrkeepsmiling)%5C **Replies:** 2\ **Last updated:** [October 17, 2026, 1:56pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbest-model-to-help-to-keep-system-in-line%2F177514 "2026-10-17T13:56:52Z")
I have been working on an AI assistant for a long time, and have tried many times to find a system for it to run on. I am a self-taught amateur, :face\_with\_peeking\_eye: but have come a long way so far. My question is ho… --- ## [Optimization query: Training efficiency on Abliterated vs. Raw checkpoints](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Foptimization-query-training-efficiency-on-abliterated-vs-raw-checkpoints%2F190690)
**Author:** [@gunnersakku](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fgunnersakku)%5C **Replies:** 1\ **Last updated:** [October 17, 2026, 10:06am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Foptimization-query-training-efficiency-on-abliterated-vs-raw-checkpoints%2F190690 "2026-10-17T10:06:13Z")
I’ve noticed a significant performance drop/inconsistency when using the abliterated version of the base model compared to the original. I’m trying to understand if the loss of the “guardrails” is causing instability in … --- ## [AI Safety is Not Ethics, It’s Plumbing: A Thermodynamic Framework for Agentic AI Loops](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fai-safety-is-not-ethics-it-s-plumbing-a-thermodynamic-framework-for-agentic-ai-loops%2F190686)
**Author:** [@RealidadAislada](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FRealidadAislada)%5C **Replies:** 0\ **Last updated:** [October 16, 2026, 9:32pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fai-safety-is-not-ethics-it-s-plumbing-a-thermodynamic-framework-for-agentic-ai-loops%2F190686 "2026-10-16T21:32:36Z")
Hi everyone, I’ve put together a two-part conceptual framework analyzing the mechanics of Agentic AI and Specification Gaming, stripping away the typical anthropomorphic tropes (“thinking minds,” “emerging consciousness… [Next page](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fc%2Fbeginners%2F5.md%3Fpage%3D1)