Replicate Intelligence #4
📰 Replicate Blog
Replicate Intelligence #4 discusses latest open-source AI models, tools, and research, including Stable Diffusion 3 Medium and real-time speech-to-text in the browser
Action Steps
- Explore the Stable Diffusion 3 Medium model and its capabilities
- Try out the real-time speech-to-text demo in the browser
- Read the research paper on extracting concepts from GPT models
Who Needs to Know This
AI engineers, data scientists, and researchers can benefit from this update on the latest AI models and tools, and use them to improve their own projects and research
Key Insight
💡 The latest AI models and tools are becoming increasingly accessible and powerful, enabling new applications and research opportunities
Share This
🤖 Latest AI updates: Stable Diffusion 3 Medium, real-time speech-to-text in the browser, and more! 📚
Key Takeaways
Replicate Intelligence #4 discusses latest open-source AI models, tools, and research, including Stable Diffusion 3 Medium and real-time speech-to-text in the browser
Full Article
# Replicate Intelligence #4 – Replicate blog
[Replicate has joined Cloudflare](https://replicate.com/blog/replicate-cloudflare)
[](https://replicate.com/)
cmd+k
[Explore](https://replicate.com/explore)[Pricing](https://replicate.com/pricing)[Enterprise](https://replicate.com/enterprise)[Docs](https://replicate.com/docs)[Blog](https://replicate.com/blog)[Sign in](https://replicate.com/signin)[Try for free](https://replicate.com/signin)
Menu
[Explore](https://replicate.com/explore)[Pricing](https://replicate.com/pricing)[Enterprise](https://replicate.com/enterprise)[Docs](https://replicate.com/docs)[Blog](https://replicate.com/blog)[Sign in](https://replicate.com/signin)[Try for free](https://replicate.com/signin)
* * *
[Compare models in the Playground](https://replicate.com/playground)
* Replicate
* [Blog](https://replicate.com/blog)
# Replicate Intelligence #4
Posted June 14, 2024 by
* [deepfates](https://replicate.com/deepfates)
Welcome to Replicate’s weekly bulletin! Each week, we’ll bring you updates on the latest open-source AI models, tools, and research. People are making cool stuff and we want to share it with you. Without further ado, here’s our hacker-in-residence deepfates with an unfiltered take on the week in AI.
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#editors-note)Editor’s note
The big open source AI news this week is the release of Stable Diffusion 3 Medium. People are already doing cool things with it, but public reaction has been mixed.
On a personal note, I got banned from X Dot Com. Apparently it is against the rules to change your profile picture to the old Twitter logo and announcing “WE ARE SO BACK”.
Anyway, here’s some things that caught my eye this week. Find me on Bluesky, I guess.
--- [deepfates](https://bsky.app/profile/deepfates.com.deepfates.com.deepfates.com.deepfates.com.deepfates.com)
* * *
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#trending-models)Trending models
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#stable-diffusion-3-medium)Stable Diffusion 3 Medium
The long-awaited image generation model is related in the 2B size (no word yet about the larger 8B version).
Users say the model is much better at creating legible text, but that it has problems with anatomy and composition.
Model weights are available under a non-commercial license.
[try on replicate](https://replicate.com/stability-ai/stable-diffusion-3)
* * *
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#cool-tools)Cool tools
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#find-concepts-in-gpt-models)Find concepts in GPT models
OpenAI does dictionary learning on their own models to extract and interpret patterns that may to specific concepts. Similar technique to the one Anthropic used to create Golden Gate Claude.
They release a research paper and feature explorer, but also code that will steer the (practically retro at this point) GPT-2-small model.
[post](https://openai.com/index/extracting-concepts-from-gpt-4/) | [paper](https://arxiv.org/abs/2406.04093) | [github](https://github.com/openai/sparse_autoencoder) | [visualizer](https://openaipublic.blob.core.windows.net/sparse-autoencoder/sae-viewer/index.html)
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#real-time-speech-to-text-in-the-browser)Real-time speech to text in the browser
The Transformers.js project has implemented OpenAI’s Whisper model in JavaScript. This means you can open a browser tab, talk to it, and get an accurate transcript of your words in real time. No coding required.
[demo](https://huggingface.co/spaces/Xenova/realtime-whisper-webgpu)
* * *
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#research-radar)Research radar
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#a-new-way-to-tokenize-images)A new way to tokenize images
Researchers at ByteDance, f
[Replicate has joined Cloudflare](https://replicate.com/blog/replicate-cloudflare)
[](https://replicate.com/)
cmd+k
[Explore](https://replicate.com/explore)[Pricing](https://replicate.com/pricing)[Enterprise](https://replicate.com/enterprise)[Docs](https://replicate.com/docs)[Blog](https://replicate.com/blog)[Sign in](https://replicate.com/signin)[Try for free](https://replicate.com/signin)
Menu
[Explore](https://replicate.com/explore)[Pricing](https://replicate.com/pricing)[Enterprise](https://replicate.com/enterprise)[Docs](https://replicate.com/docs)[Blog](https://replicate.com/blog)[Sign in](https://replicate.com/signin)[Try for free](https://replicate.com/signin)
* * *
[Compare models in the Playground](https://replicate.com/playground)
* Replicate
* [Blog](https://replicate.com/blog)
# Replicate Intelligence #4
Posted June 14, 2024 by
* [deepfates](https://replicate.com/deepfates)
Welcome to Replicate’s weekly bulletin! Each week, we’ll bring you updates on the latest open-source AI models, tools, and research. People are making cool stuff and we want to share it with you. Without further ado, here’s our hacker-in-residence deepfates with an unfiltered take on the week in AI.
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#editors-note)Editor’s note
The big open source AI news this week is the release of Stable Diffusion 3 Medium. People are already doing cool things with it, but public reaction has been mixed.
On a personal note, I got banned from X Dot Com. Apparently it is against the rules to change your profile picture to the old Twitter logo and announcing “WE ARE SO BACK”.
Anyway, here’s some things that caught my eye this week. Find me on Bluesky, I guess.
--- [deepfates](https://bsky.app/profile/deepfates.com.deepfates.com.deepfates.com.deepfates.com.deepfates.com)
* * *
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#trending-models)Trending models
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#stable-diffusion-3-medium)Stable Diffusion 3 Medium
The long-awaited image generation model is related in the 2B size (no word yet about the larger 8B version).
Users say the model is much better at creating legible text, but that it has problems with anatomy and composition.
Model weights are available under a non-commercial license.
[try on replicate](https://replicate.com/stability-ai/stable-diffusion-3)
* * *
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#cool-tools)Cool tools
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#find-concepts-in-gpt-models)Find concepts in GPT models
OpenAI does dictionary learning on their own models to extract and interpret patterns that may to specific concepts. Similar technique to the one Anthropic used to create Golden Gate Claude.
They release a research paper and feature explorer, but also code that will steer the (practically retro at this point) GPT-2-small model.
[post](https://openai.com/index/extracting-concepts-from-gpt-4/) | [paper](https://arxiv.org/abs/2406.04093) | [github](https://github.com/openai/sparse_autoencoder) | [visualizer](https://openaipublic.blob.core.windows.net/sparse-autoencoder/sae-viewer/index.html)
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#real-time-speech-to-text-in-the-browser)Real-time speech to text in the browser
The Transformers.js project has implemented OpenAI’s Whisper model in JavaScript. This means you can open a browser tab, talk to it, and get an accurate transcript of your words in real time. No coding required.
[demo](https://huggingface.co/spaces/Xenova/realtime-whisper-webgpu)
* * *
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#research-radar)Research radar
## [](https://replicate.com/blog/replicate-intelligence-2024-06-14#a-new-way-to-tokenize-images)A new way to tokenize images
Researchers at ByteDance, f
DeepCamp AI