CLIP
No written verdict yet. Here is what commenters said.
5recommend
0warn against
Dec 3, 2025first mention
Sep 24, 2026latest mention
Why people recommend it
“See CLIP https://github.com/openai/CLIP”
Show 3 more
“smaller specialised things like whisper, CLIP, SAM, etc.”
“CLIP should give the images that actually look similar.”
“Using a CLIP or Dino v2 model to produce image embeddings would probably improve the similarity search a lot”
Caveats and dissent
No one in these comments warned against it.
Often compared with
Threads where it came up
- item 46518106 2 mentions
- Contrastive Language Models 1 mention
- Getting video models to learn better, faster 1 mention
- Apple to Skip High-End M6 Mac Chips in Favor of AI-Focused M7 Line 1 mention
- Playing with Vision Embeddings 1 mention
- Ask HN: What Is the State of App Development in 2026? 1 mention
Data: clip.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 12
“CLIP[1] for actions! Very cool.”
“the hidden states of multimodal LLMs act as better conditioning mechanisms”
“the CLIP model for the context search”
“See CLIP https://github.com/openai/CLIP”
“smaller specialised things like whisper, CLIP, SAM, etc.”
“generated CLIP embeddings for all 191,922 of them”
“CLIP dominates on average”
“I'm currently exploring similar things with CLIP to guide fitness”
“CLIP should give the images that actually look similar.”
“Using a CLIP or Dino v2 model to produce image embeddings would probably improve the similarity search a lot”
“using the existing CLIP model as a visual frontend and then training a small classifier on the output of CLIP”
“normalized for CLIP embedding on GPU”