Top all-time · #380
Replicate's blog
Company blog · replicate.com/blog · on HN as bfirshaiprogramming
Replicate's blog is the company blog at replicate.com, writing about ai and programming. Since 2021, 143 different Hacker News readers have linked it 337 times, most often “Show HN: Cool site full of AI models”. It is still writing: the last post was on 2026-04-15, and you can follow it by feed.
Run open-source machine learning models with a cloud API
Most-linked posts?
- Show HN: Cool site full of AI models17 people · 20 mentions · first linked 2022-03-13
- Stable Diffusion animation7 people · 7 mentions · first linked 2022-08-29
- Run Stable Diffusion on Your M1 Mac’s GPU6 people · 6 mentions · first linked 2022-09-01
- Run Stable Diffusion with a Cloud API and Docker Image on Replicate5 people · 9 mentions · first linked 2022-08-22
- Image-to-Prompt: Get an approximate text prompt from image for Stable Diffusion5 people · 5 mentions · first linked 2022-08-25
- Fine-tune FLUX.1 with your own images5 people · 5 mentions · first linked 2024-08-15
- Recraft V3 SVG, a text-to-image model with the ability to generate SVG images5 people · 5 mentions · first linked 2024-10-30
- Fine-tune LLaMA to speak like Homer Simpson4 people · 5 mentions · first linked 2023-03-17
- Llama 3 8B and 70B pricing on replicate, also on Azure4 people · 4 mentions · first linked 2023-07-19
- /a16z-infra/llama13b-v2-chat4 people · 4 mentions · first linked 2023-07-18
What Hacker News says?
We've got an API out here: https://replicate.com/blog/run-llama-3-with-an-api You can also chat with it here: https://llama3.replicate.dev/
Comment linking /blog/run-llama-3-with-an-api
on “Meta Llama 3”Submitted Run Stable Diffusion on Your M1 Mac’s GPU as a story · 1007 points
HN discussionPublic benchmarks are broadly indicative, but devs really should run custom benchmarks on their own use cases. Replicate created a Llama 3 API [0] very quickly. This can be used to run simple benchmarks with promptfoo [1] comparing Llama 3 vs Mixtral, GPT, Claude, and others: prompts: - 'Answer this programming question concisely: {{ask}}' providers: - replicate:meta/meta-llama-3-8b-instruct - replicate:meta/meta-ll…
Comment linking /blog/run-llama-3-with-an-api
on “Meta Llama 3”Submitted Stable Diffusion animation as a story · 616 points
HN discussionBoth of those are of course out of date and require significant training instead of just feeding it a single image. InstantID ( https://replicate.com/zsxkib/instant-id ) fixes that issue.
Comment linking /zsxkib/instant-id
on “Sora is here”Submitted Snowflake Arctic Instruct (128x3B MoE), largest open source model as a story · 315 points
HN discussionHunyan ( https://replicate.com/tencent/hunyuan-video , $0.70/video) is the best but somewhat expensive. LTX ( https://replicate.com/fofr/ltx-video , $0.10) is cheaper/faster but less capable. Both are permissively licensed.
Comment linking /fofr/ltx-video
on “Sora is here”Submitted Replicate is joining Cloudflare as a story · 288 points
HN discussionFine tuning image generators with custom data sets has been done. As usual data is a big factor for the generated results latent diffusion: https://replicate.com/laion-ai/erlich vqgan + clip: https://replicate.com/ml6/julius https://www.ml6.eu/knowhow/can-ai-generate-truly-original-lo...
Comment linking /ml6/julius
on “I used DALL·E 2 to generate a logo”Submitted FLUX is fast and it's open source as a story · 258 points
HN discussionAnd it seems LAION was the “first” to offer it up for free. https://replicate.com/laion-ai/erlich
Comment linking /laion-ai/erlich
on “I used DALL·E 2 to generate a logo”Submitted FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs as a story · 228 points
HN discussionI followed https://replicate.com/blog/run-stable-diffusion-on-m1-mac
Comment linking /blog/run-stable-diffusion-on-m1-mac
on “Run Stable Diffusion on Your M1 Mac’s GPU”Submitted Animating Prompts with Stable Diffusion as a story · 158 points
HN discussionThis one is faster, I ported it https://replicate.com/kuprel/min-dalle
Comment linking /kuprel/min-dalle
on “DALL·E now available in beta”Submitted Generate images in one second on your Mac using a latent consistency model as a story · 37 points
HN discussionThere are actually multiple open source ML models for 2d to 3d which is clearly what they are doing. The difference with most of them is that this is vectors. There might actually be a similar open source model already. But I think to create it you would build it from a database of 3d assets that you could render from many angles. Probably quite similar to the way the 2d to 3d works. I don't know maybe the typical 2…
Comment linking /collections/3d-models
on “Adobe's new image rotation tool is one of the most impressive AI tools seen”Submitted Run Meta Llama 3 with an API as a story · 15 points
HN discussionAs far as I understand from a quick read of the paper, the 2D diffusion doesn't have a 3D understanding. It probably have some sort of local neighborhood understanding, aka small geometric transformation of objects map close to each other in the diffusion space (That's why like with latent spaces you can "interpolate" ( https://replicate.com/andreasjansson/stable-diffusion-animat... ) in the diffusion space). But th…
Comment linking /andreasjansson/stable-diffusion-animation
on “DreamFusion: Text-to-3D using 2D Diffusion”Submitted Cantable Diffuguesion: Bach chorale generation and harmonization as a story · 11 points
HN discussionPrevious SAM v1 you can use e.g. in here: https://fal.ai/models https://replicate.com/ You just have to wait probably few weeks for the SAM v2 to be available. Hugging Face might also have some offering
Comment linking /
on “SAM 2: Segment Anything in Images and Videos”Submitted Show HN: Cool site full of AI models as a story · 10 points
HN discussionthat's a fantastic use case!, the easiest way (and the one i'm currently using) is by upload the audio manually on : https://replicate.com/openai/whisper
Comment linking /openai/whisper
on “I record myself on audio 24x7 and use an AI to process the information”Submitted FLUX.1 Tools – Control and Steerability for Flux as a story · 7 points
HN discussionYou can use you smartphone for the recording and transcribe it here: https://replicate.com/openai/whisper regex to extract commands from the transcripted text
Comment linking /openai/whisper
on “I record myself on audio 24x7 and use an AI to process the information”Submitted How to Prompt Nano Banana Pro as a story · 6 points
HN discussionthanks!, you should try to transcribe your recordings now for free with whisper and see what you can make of them: https://replicate.com/openai/whisper
Comment linking /openai/whisper
on “I record myself on audio 24x7 and use an AI to process the information”Submitted Llama 3 8B and 70B pricing on replicate, also on Azure as a story · 6 points
HN discussionYou can run it with Cog: https://github.com/replicate/cog cog predict r8.im/stability-ai/stable-diffusion -i prompt="cow flying in space" Or you can run the Docker image directly. More details under "run on your own computer" here: https://replicate.com/stability-ai/stable-diffusion
Comment linking /stability-ai/stable-diffusion
on “Stable Diffusion Public Release”Submitted Run Stable Diffusion with a Cloud API and Docker Image on Replicate as a story · 6 points
HN discussion
