Hugging Face
The most recommended of 3 model hubs
People recommend it as the main place to find, download and run open AI models, with a large community. Main caveat: security incidents and throttled downloads.
People recommend it as the main place to find, download and run open AI models, with a large community. Main caveat: security incidents and throttled downloads.
154 people gave an opinion
- recommend it
- 92
- criticize it
- 62
- approval share
- 60%
Recommended in 12 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 92 people recommend itNeeds 10 or more, met.
- 1.4 recommenders for every criticNeeds 2 to 1 or better, not met.
- Recommended in 12 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Main place to find and download models
“I am not aware of any alternative for choosing and downloading the right model for local inference that is superior to using directly the Huggingface site”
Comment from 16 Apr 2026 in “Stop Using Ollama” “I went with Hugging Face mainly because it is simple and already handles distribution and versioning”
Comment from 19 Mar 2026 in “Show HN: Hacker News archive (47M+ items, 11.6GB) as Parquet, updated every 5m” “One good way to find inference providers for open models is through hugging face”
Comment from 31 Jul 2026 in “Ask HN: What are you using for LLM inference in production?” - Comment from 9 Feb 2026 in “No one on Hacker News can answer this question”
- Comment from 27 Aug 2026 in “Small Models Have Arrived”
- Comment from 28 Nov 2025 in “A trillion dollars (potentially) wasted on gen-AI”
Trusted open model community
“HF has become the front runner in open model hosting. They have a lot of trust from the wider community.”
Comment from 27 Aug 2026 in “Nvidia agrees to acquire Hugging Face for $13B” “nothing else has been as effective at accelerating model development and research sharing”
Comment from 11 Sep 2026 in “HuggingFace: Security.txt” “building such a huge AI community around a single place is probably impossible by now”
Comment from 27 Aug 2026 in “Nvidia agrees to acquire Hugging Face for $13B” Free or cheap model hosting
“Freemium seems to be working pretty well for them—what's the alternative website, after all. They seem to command their niche.”
Comment from 20 Feb 2026 in “Ggml.ai joins Hugging Face to ensure the long-term progress of Local AI” “there's places (hugging face?) that will host and run your updated model for cheaper than you could run it yourself”
Comment from 22 Dec 2025 on Hacker News “who already did all the setup for you and just waits for your credit card per minute/seconds/token”
Comment from 16 Apr 2026 in “Darkbloom – Private inference on idle Macs” Useful libraries and tooling
“they also have training, inference hosting, and they develop/maintain a bunch of core AI infrastructure software”
Comment from 27 Aug 2026 in “Nvidia agrees to acquire Hugging Face for $13B” “HF is super valuable with their transformers library, their code, tutorials, smol-models, etc”
Comment from 20 Feb 2026 in “Ggml.ai joins Hugging Face to ensure the long-term progress of Local AI” “Getting stuff to run before huggingface and uv and docker containers with cuda was way worse.”
Comment from 10 Mar 2026 in “Open Weights Isn't Open Training”
What people warn about
Each mark is one comment
Security incidents and hacks
“It is extremely embarrassing to get owned because you are letting arbitrary programs/users call out to the open web from the infra”
Comment from 15 Sep 2026 in “Dario, Please” “If you think otherwise, look no further than the Huggingface hack and what happens when they run completely autonomously.”
Comment from 16 Oct 2026 in “Ask HN: Thoughts on hiring junior engineers in 2026?” “HuggingFace wasn't doing at all while also allowing any user's arbitrary programs to call out to the open web from prod”
Comment from 18 Sep 2026 in “A heap overflow and SSO misconfiguration to compromise OpenAI internal repos” Lock-in and control after acquisitions
“This acquisition will likely solidify this general stance, reduce free compute allowance for a subscription, cap downloads”
Comment from 27 Aug 2026 in “Nvidia agrees to acquire Hugging Face for $13B” “now the company that sells the GPUs also owns the registry that decides which models are easy to find and download”
Comment from 28 Aug 2026 in “Nvidia agrees to acquire Hugging Face for $13B” “Combined with recent acquisition of OpenRouter we see exactly what we were increasingly concerned about”
Comment from 27 Aug 2026 in “Nvidia agrees to acquire Hugging Face for $13B” - Comment from 21 Sep 2026 in “Nobody pays for FOSS, we can force them to”
Gated downloads, throttling and telemetry
“an onerous license agreement that must be agreed to during registration and a bloated cli interface with loads of telemetry”
Comment from 4 Sep 2026 in “Georgi Gerganov on llama.cpp/ggml future after Nvidia acquisition of HuggingFace” “Previously with non-gated models, you could download the XET files with any program, no special software needed.”
Comment from 16 Oct 2026 in “Muse Spark Open” “pulling models is an absolute nightmare when I basically have to use HF and even with a token getting throttled”
Comment from 27 Aug 2026 in “Nvidia agrees to acquire Hugging Face for $13B” Inference hosting feels unserious
“Their inference hosting is a joke (I don't think I ever had a working response from a "try this model" form).”
Comment from 27 Aug 2026 in “Nvidia agrees to acquire Hugging Face for $13B” “Huggingface for a serious service has never felt truly serious to me for reasons like these.”
Comment from 1 Aug 2026 in “Tailscale didn't stop the Hugging Face intrusion”
Other model hubs people recommend
The two most recommended other model hubs, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Hugging FaceThis page | 92 | 62 | 60% | 12 | — | — |
| ModelScope | 9 | 0 | 100% | 5 | — | — |
| Civitai | 7 | 5 | 58% | 5 | — | — |
What people said
40 of the 163 comments that mention it, newest first. Each links to the original comment.
“If you think otherwise, look no further than the Huggingface hack and what happens when they run completely autonomously.”
“Previously with non-gated models, you could download the XET files with any program, no special software needed.”
“if not for arxiv and huggingface, the field of ML would be nowhere near where it is today”
“How they host stuff and then leverage that position to build add-ons and lock-in, lobby for laws.”
“HuggingFace wasn't doing at all while also allowing any user's arbitrary programs to call out to the open web from prod”
“It is extremely embarrassing to get owned because you are letting arbitrary programs/users call out to the open web from the infra”
“nothing else has been as effective at accelerating model development and research sharing”
“an onerous license agreement that must be agreed to during registration and a bloated cli interface with loads of telemetry”
“use 3rd party marketplaces. cerebras is resold on vercel, openrouter and huggingface.”
“now the company that sells the GPUs also owns the registry that decides which models are easy to find and download”
“providing a sufficiently useful and general service that makes technology more accessible to all”
“I hooked up the settings to download models from Hugging Face conveniently”
“HF has become the front runner in open model hosting. They have a lot of trust from the wider community.”
“Their inference hosting is a joke (I don't think I ever had a working response from a "try this model" form).”
“HF has been a close part of my ML/AI career, coinciding exactly when I moved into this space 10 years ago.”
“building such a huge AI community around a single place is probably impossible by now”
“Combined with recent acquisition of OpenRouter we see exactly what we were increasingly concerned about”
“This acquisition will likely solidify this general stance, reduce free compute allowance for a subscription, cap downloads”
“pulling models is an absolute nightmare when I basically have to use HF and even with a token getting throttled”
“Brilliant replacement of http://build.nvidia.com to increase developer reach and deployment of nvidia open models”
“they also have training, inference hosting, and they develop/maintain a bunch of core AI infrastructure software”
“the pricing of tokens on places like HuggingFace and OpenRouter suggests that probably isn't true”
“Huggingface for a serious service has never felt truly serious to me for reasons like these.”
“One good way to find inference providers for open models is through hugging face”
“I am not aware of any alternative for choosing and downloading the right model for local inference that is superior to using directly the Huggingface site”
“who already did all the setup for you and just waits for your credit card per minute/seconds/token”
“Huggingface, Swartz et al have done more social/political good for this world than billions have.”
“The red/yellow/green indicators on huggingface.co are kind of nice, but you only know for sure when you try to load the model”
“I went with Hugging Face mainly because it is simple and already handles distribution and versioning”
“Getting stuff to run before huggingface and uv and docker containers with cuda was way worse.”
“The other nice part of huggingface's setup is you can add theoretical hardware and search that way too.”
“And I said all of that despite us continuing to use their platform and libraries extensively…”
“Freemium seems to be working pretty well for them—what's the alternative website, after all. They seem to command their niche.”
“HF is super valuable with their transformers library, their code, tutorials, smol-models, etc”
“one of the few quiet heroes (along with Chinese OSS) helping bring on-premise AI to the masses”
“Hugging Face has large number of coding models that you can download for free”
“benchmark the models you could run on your potential hardware on (neo) cloud provider like HuggingFace”
“there's places (hugging face?) that will host and run your updated model for cheaper than you could run it yourself”
“There are plenty of models on HuggingFace that I can run on my Mac Studio”
“All the data used by Apertus is just data processed or generated by American companies(NVidia, Apple and huggingface mostly).”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.