Hassan's banner

Hassan

@nutlope • 102,743 subscribers

Head of Developer Experience @togethercompute. Building open source AI apps like https://t.co/f8hbvXOFaN, https://t.co/SmHisRTtnp, & https://t.co/hs63SlHxjw.

Shorts

I turned my "read later" list into a personal library

I turned my "read later" list into a personal library

169,254 просмотров

How I read articles now

How I read articles now

135,975 просмотров

This model is insane at design. I asked GLM 5.2 (left) and Opus 4.8 (right) to build me a landing page and you can't even tell the difference. GLM cost $0.06 while opus cost $0.49. More than 6x cheaper while being faster + more token efficient. Another win for open source AI.

This model is insane at design. I asked GLM 5.2 (left) and Opus 4.8 (right) to build me a landing page and you can't even tell the difference. GLM cost $0.06 while opus cost $0.49. More than 6x cheaper while being faster + more token efficient. Another win for open source AI.

1,583,619 просмотров

I used Jev to classify 1,018 AI research papers. The result: $0.08 total cost and 256ms median end-to-end latency per paper. The pipeline was: 1. Summarize each paper with DeepSeek V4 Flash 2. Send the title + summary + 24 possible topics to Jev 3. Use Jev to classify each paper 4. Visualize everything on The summaries cost $3.99 on Together AI. The classifications cost $0.08 on TypeSafe AI. So for just over $4 of inference, I ended up with a pretty useful way to explore the top AI research papers from the past year. I think this is where things are heading: different models for different parts of the workflow, instead of using one model for everything. I’m running evals on the Jev classifications before replacing the current ones, but the site is already live:

I used Jev to classify 1,018 AI research papers. The result: $0.08 total cost and 256ms median end-to-end latency per paper. The pipeline was: 1. Summarize each paper with DeepSeek V4 Flash 2. Send the title + summary + 24 possible topics to Jev 3. Use Jev to classify each paper 4. Visualize everything on The summaries cost $3.99 on Together AI. The classifications cost $0.08 on TypeSafe AI. So for just over $4 of inference, I ended up with a pretty useful way to explore the top AI research papers from the past year. I think this is where things are heading: different models for different parts of the workflow, instead of using one model for everything. I’m running evals on the Jev classifications before replacing the current ones, but the site is already live:

341,359 просмотров

What if your X bookmarks + saved articles could become a physical book? Save articles from anywhere, read them in a personal library, and turn your favorites into a book or podcast. Gloss launches in 24 hours. Gloss Print + Podcast coming soon after. Free and open source!

What if your X bookmarks + saved articles could become a physical book? Save articles from anywhere, read them in a personal library, and turn your favorites into a book or podcast. Gloss launches in 24 hours. Gloss Print + Podcast coming soon after. Free and open source!

27,483 просмотров

Introducing I took the top 1k research papers of the last year, summarized them, and visualized them. Fun fact: all 1,000 papers cost a total of $4 to summarize with DeepSeek V4 Flash.

Introducing I took the top 1k research papers of the last year, summarized them, and visualized them. Fun fact: all 1,000 papers cost a total of $4 to summarize with DeepSeek V4 Flash.

134,313 просмотров

This model is insane at 3D website design. I asked GLM 5.3 (left) & Fable 5 (right) to make me a 3d biking website. GLM nailed it while Fable failed. GLM cost $0.14 while Fable cost $2.21. More than 15x cheaper while being even better in this case!

This model is insane at 3D website design. I asked GLM 5.3 (left) & Fable 5 (right) to make me a 3d biking website. GLM nailed it while Fable failed. GLM cost $0.14 while Fable cost $2.21. More than 15x cheaper while being even better in this case!

119,361 просмотров

Jev + Kimi K3 for fraud detection! TLDR: Jev classified 100 emails in 1.42 seconds, then I routed the uncertain cases to Kimi K3. The full pipeline got 96/100 correct for only ~$0.07. Video is not sped up, check out the live run! Here was my process: I gave Jev 100 emails to classify (a mix of 50 legit & 50 fraudelent emails). It classified all of them in 1.42 seconds. An underrated feature about Jev is it will give you the confidence score for a classification, so I routed any prediction under 95% confidence to Kimi K3 to be fully sure. 31 emails fell below that threshold. After routing those to Kimi K3, the combined pipeline reached 96% accuracy. The full run took 16 seconds & ~$0.07 in inference costs: - $0.068 from Kimi K3 on Together AI - $0.003 (1/3 of a cent) from Jev on TypeSafe AI. I think this is a really interesting pattern: use a fast specialized model like Jev for the narrow task, then route the uncertain cases to a larger LLM. I feel like this kind of approach could be a game changer for use cases like fraud or anything realtime. You can use the speed & low cost of Jev while having a larger LLM as a fallback to ensure high accuracy.

Jev + Kimi K3 for fraud detection! TLDR: Jev classified 100 emails in 1.42 seconds, then I routed the uncertain cases to Kimi K3. The full pipeline got 96/100 correct for only ~$0.07. Video is not sped up, check out the live run! Here was my process: I gave Jev 100 emails to classify (a mix of 50 legit & 50 fraudelent emails). It classified all of them in 1.42 seconds. An underrated feature about Jev is it will give you the confidence score for a classification, so I routed any prediction under 95% confidence to Kimi K3 to be fully sure. 31 emails fell below that threshold. After routing those to Kimi K3, the combined pipeline reached 96% accuracy. The full run took 16 seconds & ~$0.07 in inference costs: - $0.068 from Kimi K3 on Together AI - $0.003 (1/3 of a cent) from Jev on TypeSafe AI. I think this is a really interesting pattern: use a fast specialized model like Jev for the narrow task, then route the uncertain cases to a larger LLM. I feel like this kind of approach could be a game changer for use cases like fraud or anything realtime. You can use the speed & low cost of Jev while having a larger LLM as a fallback to ensure high accuracy.

60,975 просмотров

Announcing the NYC AI Atlas! An open source 3D map of the top AI startups in New York City.

Announcing the NYC AI Atlas! An open source 3D map of the top AI startups in New York City.

106,053 просмотров

This model is insane at landing pages. I asked DeepSeek V4.1 Flash & Claude Fable 5 to build me a landing page for a movie theater. Fable cost $1.21 while V4.1 Flash cost 2.6 cents, making it more than 40x cheaper at similar quality. Gave both the exact same prompt!

This model is insane at landing pages. I asked DeepSeek V4.1 Flash & Claude Fable 5 to build me a landing page for a movie theater. Fable cost $1.21 while V4.1 Flash cost 2.6 cents, making it more than 40x cheaper at similar quality. Gave both the exact same prompt!

55,618 просмотров

Jev vs GLM 5.3 at chess! Results: ◾ GLM 5.3 won by checkmate in 29 moves ◾ Jev: ~0.3s and <$0.0001 per move ◾ GLM 5.3: ~5.8s and ~$0.008 per move ◾ The whole game cost 24 cents My main takeaway is that it's often useful to use each one to their strengths: ◾ Fast, well-defined classification → specialized models like Jev ◾ Classifications that need reasoning or lookahead → LLMs like GLM 5.3 ◾ Real classification pipelines → hybrid. Jev handles the easy calls, an open model handles the hard ones. The future is multi-model!

Jev vs GLM 5.3 at chess! Results: ◾ GLM 5.3 won by checkmate in 29 moves ◾ Jev: ~0.3s and <$0.0001 per move ◾ GLM 5.3: ~5.8s and ~$0.008 per move ◾ The whole game cost 24 cents My main takeaway is that it's often useful to use each one to their strengths: ◾ Fast, well-defined classification → specialized models like Jev ◾ Classifications that need reasoning or lookahead → LLMs like GLM 5.3 ◾ Real classification pipelines → hybrid. Jev handles the easy calls, an open model handles the hard ones. The future is multi-model!

39,435 просмотров

Announcing LogoCreator v2! Create professional logos & brand images in seconds. Free and 100% open source →

86,190 просмотров

Announcing Face Photo Restorer! Restore any old blurry face photos with AI – 100% free and open source.

Announcing Face Photo Restorer! Restore any old blurry face photos with AI – 100% free and open source.

1,180,315 просмотров

Announcing llama-ocr – a free + open source OCR tool! It takes documents (images for now) & outputs markdown, and does really well for complex receipts, PDFs with tables/charts, ect... Powered by Llama 3.2 vision on Together AI & available on npm today!

Announcing llama-ocr – a free + open source OCR tool! It takes documents (images for now) & outputs markdown, and does really well for complex receipts, PDFs with tables/charts, ect... Powered by Llama 3.2 vision on Together AI & available on npm today!

444,290 просмотров

Working on a new skill where Fable acts as an orchestrator and delegates work to OpenCode subagents running OSS models. It delegates to GLM 5.3 Flash for easier tasks and Kimi K3 for harder tasks, all running on Together AI. GLM 5.3 Flash is 100x cheaper than Fable for output tokens while being pretty good at coding so I'm excited to experiment with how well this works.

Working on a new skill where Fable acts as an orchestrator and delegates work to OpenCode subagents running OSS models. It delegates to GLM 5.3 Flash for easier tasks and Kimi K3 for harder tasks, all running on Together AI. GLM 5.3 Flash is 100x cheaper than Fable for output tokens while being pretty good at coding so I'm excited to experiment with how well this works.

35,379 просмотров

Announcing GLM Arena! A series of tests (infographics, svgs, sites, ect..) ran on GLM 5.2 and Opus 4.8, with prompts included. On average, GLM 5.2 produced 2x the tokens but was still faster + 3x cheaper with similar quality.

Announcing GLM Arena! A series of tests (infographics, svgs, sites, ect..) ran on GLM 5.2 and Opus 4.8, with prompts included. On average, GLM 5.2 produced 2x the tokens but was still faster + 3x cheaper with similar quality.

75,421 просмотров

Ran 10 more tests comparing GLM 5.2 & Opus. On average, GLM 5.2 produced 2x the tokens but was still faster + 3x cheaper with similar quality! I'm open sourcing all these tests tomorrow, including the code, my prompts, and the token/cost stats.

Ran 10 more tests comparing GLM 5.2 & Opus. On average, GLM 5.2 produced 2x the tokens but was still faster + 3x cheaper with similar quality! I'm open sourcing all these tests tomorrow, including the code, my prompts, and the token/cost stats.

67,027 просмотров

I built a design MCP server for Claude Code, Codex, and OpenCode. It searches hundreds of beautiful websites and gives your coding agent design inspiration. Launching next week.

I built a design MCP server for Claude Code, Codex, and OpenCode. It searches hundreds of beautiful websites and gives your coding agent design inspiration. Launching next week.

28,328 просмотров

Building an app to help people track their finances! It'll take in any receipts you have, extract the total $, and categorize it for you (dining, groceries, utilities, ect). Launching soon. Will be 100% free & open source.

Building an app to help people track their finances! It'll take in any receipts you have, extract the total $, and categorize it for you (dining, groceries, utilities, ect). Launching soon. Will be 100% free & open source.

138,237 просмотров

I built a menu visualizer app that can take a restaurant menu pic and generate nice images for each dish! 100% free and open source. Dropping in 24 hours.

I built a menu visualizer app that can take a restaurant menu pic and generate nice images for each dish! 100% free and open source. Dropping in 24 hours.

166,156 просмотров

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Introducing Gloss! A beautiful new way to save, read, and interact with articles. → Save articles, PDFs, and X bookmarks → Highlight and get instant answers with AI → Build a personal library you'll actually use Free and open source!

Introducing Gloss! A beautiful new way to save, read, and interact with articles. → Save articles, PDFs, and X bookmarks → Highlight and get instant answers with AI → Build a personal library you'll actually use Free and open source!

77,552 просмотров • 3 дней назад

Turn your X bookmarks into a personal library with Grok Bot! Launching in 48 hours. Will be 100% free & open source. I'm sending out a few more early access invites for people who want to try it out and give feedback. Reply if you're interested!

Turn your X bookmarks into a personal library with Grok Bot! Launching in 48 hours. Will be 100% free & open source. I'm sending out a few more early access invites for people who want to try it out and give feedback. Reply if you're interested!

98,732 просмотров • 6 дней назад

I got tired of CleanShot, so I built my own Mac screenshot app. It's called Matte. More backgrounds, better customization, automatic hiding of sensitive info, and all the little features I personally wanted. We're entering the age of personal software.

I got tired of CleanShot, so I built my own Mac screenshot app. It's called Matte. More backgrounds, better customization, automatic hiding of sensitive info, and all the little features I personally wanted. We're entering the age of personal software.

59,265 просмотров • 4 дней назад

Introducing Inspo. A design MCP server for Claude Code, Codex, and OpenCode. It searches 800+ beautiful websites and finds relevant design inspiration for your coding agent. Install → npx inspo-mcp install

Introducing Inspo. A design MCP server for Claude Code, Codex, and OpenCode. It searches 800+ beautiful websites and finds relevant design inspiration for your coding agent. Install → npx inspo-mcp install

415,924 просмотров • 1 месяц назад

Introducing Hallmark! An open source design skill to make beautiful UIs and landing pages by default. Works in Claude Code, Cursor, and Codex. npx skills add nutlope/hallmark

Introducing Hallmark! An open source design skill to make beautiful UIs and landing pages by default. Works in Claude Code, Cursor, and Codex. npx skills add nutlope/hallmark

852,861 просмотров • 5 месяцев назад

Since this went a bit viral, here's a 1 minute demo! Launching soon. 100% free & open source. Gonna send out some early access invites this weekend. Lmk if you wanna try it out & send feedback!

Since this went a bit viral, here's a 1 minute demo! Launching soon. 100% free & open source. Gonna send out some early access invites this weekend. Lmk if you wanna try it out & send feedback!

31,145 просмотров • 7 дней назад

Announcing roomGPT! Redesign your room in seconds with AI! 100% free and open source.

Announcing roomGPT! Redesign your room in seconds with AI! 100% free and open source.

1,479,999 просмотров • 3 лет назад

Stop letting your agents ship ugly UIs. In this talk, I break down everything I know about how to make AI apps with great designs:

Stop letting your agents ship ugly UIs. In this talk, I break down everything I know about how to make AI apps with great designs:

78,824 просмотров • 1 месяц назад

Built an app that can generate an entire comic in a few seconds. You can upload your own characters, choose a theme, & build your own comic page by page! 100% free and open source. Launching in 24 hours.

Built an app that can generate an entire comic in a few seconds. You can upload your own characters, choose a theme, & build your own comic page by page! 100% free and open source. Launching in 24 hours.

253,970 просмотров • 8 месяцев назад

Building an app that can generate infographics from PDFs! Powered by Nano Banana Pro & coming soon.

Building an app that can generate infographics from PDFs! Powered by Nano Banana Pro & coming soon.

281,117 просмотров • 10 месяцев назад

BillSplit – the easiest way to split any restaurant bill – is dropping in 48 hours. Will be 100% free & open source!

BillSplit – the easiest way to split any restaurant bill – is dropping in 48 hours. Will be 100% free & open source!

350,498 просмотров • 1 год назад

Introducing LlamaCoder! An open source Claude Artifacts app that can generate full React apps and components with Llama 3.1 405B. 100% free and open source.

Introducing LlamaCoder! An open source Claude Artifacts app that can generate full React apps and components with Llama 3.1 405B. 100% free and open source.

501,759 просмотров • 2 лет назад

Announcing SubStudio! Generate subtitles for any video in seconds with AI. 100% free & open source! Powered by Whisper on Together AI and FFmpeg via fluent-ffmpeg.

Announcing SubStudio! Generate subtitles for any video in seconds with AI. 100% free & open source! Powered by Whisper on Together AI and FFmpeg via fluent-ffmpeg.

136,421 просмотров • 6 месяцев назад

Introducing Whisper – an open source voice note taking app! Record voice notes and transcribe them into lists, blogs, & more with AI. 100% free & open source.

Introducing Whisper – an open source voice note taking app! Record voice notes and transcribe them into lists, blogs, & more with AI. 100% free & open source.

281,215 просмотров • 1 год назад

Introducing LogoCreator! An open source logo generator that creates professional logos in seconds using Flux Pro 1.1 on Together AI. 100% free and open source. Demo + code:

351,715 просмотров • 1 год назад

Introducing Figure out which open model is best for your use case. Compare models across coding, agents, long context, vision, finance, and more. Then see how they compare on cost + quality, including what you could save by moving to open models.

Introducing Figure out which open model is best for your use case. Compare models across coding, agents, long context, vision, finance, and more. Then see how they compare on cost + quality, including what you could save by moving to open models.

18,655 просмотров • 25 дней назад

Watch GLM 5.3 add a new page on my personal site in 37 seconds. The video is not sped up. GLM 5.3 at 200+ TPS, live on the Together AI API.

Watch GLM 5.3 add a new page on my personal site in 37 seconds. The video is not sped up. GLM 5.3 at 200+ TPS, live on the Together AI API.

30,322 просмотров • 1 месяц назад

Opus 5.5 is incredible at launch videos. I gave it an app I launched recently & it one-shotted a great launch video for me. Was very impressed with the animations & charts. I love open models and use them a lot, but I'm also a big believer in using the right model for the job

Opus 5.5 is incredible at launch videos. I gave it an app I launched recently & it one-shotted a great launch video for me. Was very impressed with the animations & charts. I love open models and use them a lot, but I'm also a big believer in using the right model for the job

16,453 просмотров • 22 дней назад

My free + open source logo generator is dropping in 24h!

316,828 просмотров • 1 год назад

Announcing Agent Recipes! A site to learn about agent/workflow recipes with code examples that you can easily copy & paste into your own AI apps. I'm gonna make this the go-to resource for devs to learn about agents & how to implement them – more soon.

Announcing Agent Recipes! A site to learn about agent/workflow recipes with code examples that you can easily copy & paste into your own AI apps. I'm gonna make this the go-to resource for devs to learn about agents & how to implement them – more soon.

256,250 просмотров • 1 год назад