Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Prompting GPT-5 is different. In the examples below, optimized prompts: • Cut runtime by 1s • Dropped memory use 3,626 KB → 577 KB • Boosted code quality • Improved robustness (0.32→0.54) • Increased context grounding (0.80→0.95) We built a prompt migrator + optimizer so you don’t need to... show more

edwin

30,629 subscribers

269,224 Aufrufe • vor 11 Monaten •via X (Twitter)

Wissenschaft & Technologie Bildung

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

.OpenAI's GPT-5.1 is live in Cline. After testing the last several weeks, here's what we've found makes GPT-5.1 an exceptional model in Cline: GPT-5.1 is an obsessive researcher. It reads everything before writing anything. Where other models grab context and go, GPT-5.1 investigates thoroughly first, even for one-line changes. We found this overkill for simple tasks, and modified GPT-5.1's prompting resultantly, but a superpower when channeled right. Here's what you should expect from GPT-5.1 in Cline: > deep, systematic investigation before any edits > follows constraints religiously (specs, todo lists, your requirements) > maintains coherent reasoning with large contexts > delivers complete features with fewer corrections How we optimized Cline for GPT-5.1: 1. redesigned prompts for execution focus and stricter Plan/Act transitions 2. tightened tool specs, GPT-5.1 internalizes rules with just single examples 3. split /deep-planning investigation into two phases (read first, then search) Result: GPT-5.1 is a model that finishes what it starts, even across dozens of files and thousands of lines. Both GPT-5.1 (272K context) and GPT-5.1-codex (400K context) available now in Cline. $1.25/M input, $10/M output. Enable native tool calling in settings for best results. Use /deep-planning to see GPT-5.1 at its best. Full technical breakdown linked below.

.OpenAI's GPT-5.1 is live in Cline. After testing the last several weeks, here's what we've found makes GPT-5.1 an exceptional model in Cline: GPT-5.1 is an obsessive researcher. It reads everything before writing anything. Where other models grab context and go, GPT-5.1 investigates thoroughly first, even for one-line changes. We found this overkill for simple tasks, and modified GPT-5.1's prompting resultantly, but a superpower when channeled right. Here's what you should expect from GPT-5.1 in Cline: > deep, systematic investigation before any edits > follows constraints religiously (specs, todo lists, your requirements) > maintains coherent reasoning with large contexts > delivers complete features with fewer corrections How we optimized Cline for GPT-5.1: 1. redesigned prompts for execution focus and stricter Plan/Act transitions 2. tightened tool specs, GPT-5.1 internalizes rules with just single examples 3. split /deep-planning investigation into two phases (read first, then search) Result: GPT-5.1 is a model that finishes what it starts, even across dozens of files and thousands of lines. Both GPT-5.1 (272K context) and GPT-5.1-codex (400K context) available now in Cline. $1.25/M input, $10/M output. Enable native tool calling in settings for best results. Use /deep-planning to see GPT-5.1 at its best. Full technical breakdown linked below.

Cline

72,111 Aufrufe • vor 8 Monaten

Claude Opus 5 just one-shotted this FPS. Less than 1 hour from prompt to playable demo. Every weapon, enemy, mechanic and environment was generated from scratch. Prompt below 👇 There is no prompt below, because the first section is exactly the kind of post I’m tired of seeing every time a new model drops. A flashy demo, “one-shot” in the caption, “built in under an hour”… anything to farm impressions while conveniently hiding the actual process. If it was really made from a single prompt, show the prompt. That’s what I do on this account. When I claim something is a one-prompt prototype built with GPT-5.6 Sol Ultra and Codex, I share the prompt publicly so anyone can verify it. AI is already making game prototyping incredible. We don’t need fake “one-shot” stories to prove it. Original footage:

Claude Opus 5 just one-shotted this FPS. Less than 1 hour from prompt to playable demo. Every weapon, enemy, mechanic and environment was generated from scratch. Prompt below 👇 There is no prompt below, because the first section is exactly the kind of post I’m tired of seeing every time a new model drops. A flashy demo, “one-shot” in the caption, “built in under an hour”… anything to farm impressions while conveniently hiding the actual process. If it was really made from a single prompt, show the prompt. That’s what I do on this account. When I claim something is a one-prompt prototype built with GPT-5.6 Sol Ultra and Codex, I share the prompt publicly so anyone can verify it. AI is already making game prototyping incredible. We don’t need fake “one-shot” stories to prove it. Original footage:

Givros

14,794 Aufrufe • vor 8 Tagen

Fable 5 and GPT-5.6 built the same scroll-animated website from one skill in 32 minutes. Only one of them made it feel like a film. Same prompt. Boutique Japan travel brand, origami style. A subway pulls in, a paper house unfolds into a hotel, a bird takes flight as you scroll. Doing this by hand is brutal, so the skill handles everything automatically. Here is how it works: → One anchor image sets the visual style and every scene that follows inherits it automatically → Higgsfield generates the video straight from the terminal through the MCP connection, no manual exporting → FFmpeg rips every frame out of the generated video automatically → Each frame then gets mapped to a scroll position so your scrollbar literally becomes the playhead The verdict came down to transitions. GPT-5.6 built strong individual scenes with incredible detail inside each one, but then hard-cut between them with no connection. Fable 5 stitched everything together. Wires push out of the top of the frame while the next scene blurs in behind it, gains depth and locks into place seamlessly. Same skill, same prompts, 10 generations each from both models. GPT-5.6 built 4 scenes. Fable 5 built one continuous film. Credit: Chase AI

0xMarioNawfal

48,091 Aufrufe • vor 21 Tagen

MASSIVE Hermes Agent update over the last few days Totally changes the way I use Hermes Here's 6 new features you need to start using immediately (video demoing them below): 1. Mixture of agents: send your prompt to a team of different models. The team sends back all of their responses to an orchestrator model who synthesizes a final answer. Gives much better results than just sending a prompt to 1 model 2. /learn: use the new built in /learn skill to have Hermes automatically create new skills. You can either give a prompt after /learn or put in a URL. I like pasting in URLs of tweets with helpful tips after /learn and Hermes will automatically turn it into a skill 3. /journey: See every skill and memory Hermes has created for you on a really nice timeline/chart. Great for seeing how your agent has learned and improved over time 4. Self improvement cost savings: Hermes now uses cheaper models to do it's self improvement including memory creation and skill creation. These types of activities happen in the background of almost every prompt, so this results in TONS of cost savings over time 5. Vibe coding improvements: Hermes desktop is now a full vibe coding tool. You can see diffs, make commits, and even open up PRs directly from the desktop interface. Makes it WAYYY nicer to vibe code with 6. Fable 5 is now built in. Fable 5. Obviously Fable is incredibly expensive, so only use this new profile for incredibly complex tasks. Excellent updates that have significantly improved the experience. Video demoing all the updates below!

Alex Finn

143,012 Aufrufe • vor 27 Tagen

so much of work is what I like to call the “work around work” - switching between systems, chasing context, and stitching tools together to get things done company knowledge in chatgpt now does this for you it can reason across tools like slack, sharepoint, google drive, github, and more to get answers, do analysis, and take action using a version of gpt-5 that’s optimized for working with tools & providing citations company knowledge has changed how i use chatgpt at work more than anything we have built so far - let us know what you think!

so much of work is what I like to call the “work around work” - switching between systems, chasing context, and stitching tools together to get things done company knowledge in chatgpt now does this for you it can reason across tools like slack, sharepoint, google drive, github, and more to get answers, do analysis, and take action using a version of gpt-5 that’s optimized for working with tools & providing citations company knowledge has changed how i use chatgpt at work more than anything we have built so far - let us know what you think!

Brad Lightcap

315,342 Aufrufe • vor 9 Monaten

THIS GUY SPENT 5 MINUTES SETTING UP A $20 CLAUDE WORKSPACE AND TURNED 30 MINUTES OF DAILY PROMPT CHAOS INTO A SYSTEM Most people open Claude, type 1 messy prompt, rewrite it 10 times, get the same generic answer and call the model “bad.” He opens Claude Cowork, adds the project rules, past context, working files, examples, decisions and a clean place for the model to remember what it is doing. Same Claude. Same $20 tool. Completely different output. The funny part is there is no magic prompt here. No secret template. No “10x Claude” trick. Just the boring layer Karpathy keeps pointing at: context, memory and structure before output. A setup like this can save 30 minutes a day, which is 15 hours a month. If you use Claude for client work at $50 to $100/hour, that is $750 to $1,500/month in time you stop burning on repeated prompts. The model is real. The output is real. The edge is in not making Claude start from zero every single time.

THIS GUY SPENT 5 MINUTES SETTING UP A $20 CLAUDE WORKSPACE AND TURNED 30 MINUTES OF DAILY PROMPT CHAOS INTO A SYSTEM Most people open Claude, type 1 messy prompt, rewrite it 10 times, get the same generic answer and call the model “bad.” He opens Claude Cowork, adds the project rules, past context, working files, examples, decisions and a clean place for the model to remember what it is doing. Same Claude. Same $20 tool. Completely different output. The funny part is there is no magic prompt here. No secret template. No “10x Claude” trick. Just the boring layer Karpathy keeps pointing at: context, memory and structure before output. A setup like this can save 30 minutes a day, which is 15 hours a month. If you use Claude for client work at $50 to $100/hour, that is $750 to $1,500/month in time you stop burning on repeated prompts. The model is real. The output is real. The edge is in not making Claude start from zero every single time.

Gipp 🦅

14,871 Aufrufe • vor 2 Monaten

Air Canada’s chatbot once literally made up its own refund policy in court and won a lawsuit for the customer, not the airline. There’s a new term being coined right now called vibe training by the company Plurai, and they’ve basically built a way to use tiny, fast models as guardrails to catch hallucinations in sub-100ms and the cost is over 8x lower than GPT-5-mini. 🔥👉 They’re live on Product Hunt today: If you’re building agents, go check them out, grab the free trial, and show them some love on the launch! 🫶 The best part? You don’t need a PhD in AI. Sponsored by Plurai.

Air Canada’s chatbot once literally made up its own refund policy in court and won a lawsuit for the customer, not the airline. There’s a new term being coined right now called vibe training by the company Plurai, and they’ve basically built a way to use tiny, fast models as guardrails to catch hallucinations in sub-100ms and the cost is over 8x lower than GPT-5-mini. 🔥👉 They’re live on Product Hunt today: If you’re building agents, go check them out, grab the free trial, and show them some love on the launch! 🫶 The best part? You don’t need a PhD in AI. Sponsored by Plurai.

Kunal Kushwaha

11,959 Aufrufe • vor 3 Monaten

New short course: DSPy: Build and Optimize Agentic Apps DSPy is a powerful open-source framework for automatically tuning prompts for GenAI applications. In this course, you'll learn to use DSPy, together with MLflow. This is built in partnership with Databricks and taught by Chen Qian, co-lead of the DSPy framework. Many AI builders spend hours hand-tuning prompts. When given a set of evals, DSPy automates this process. It’s especially useful for optimizing prompts, including few-shot prompts, in complex agentic AI workflows. Further, if you switch an application to a newer LLM, performance can degrade if your prompts were optimized to the previous model. DSPy automatically optimizes the entire system for the new LLM as well, using just a few evaluation examples. This course teaches DSPy works, and best practices for using it. You’ll write programs using DSPy’s signature-based programming model, debug them with MLflow tracing -- to gain visibility into how different parts of a pipeline, as well as how the overall system, are performing -- and automatically improve their accuracy with DSPy Optimizer. Please sign up here:

New short course: DSPy: Build and Optimize Agentic Apps DSPy is a powerful open-source framework for automatically tuning prompts for GenAI applications. In this course, you'll learn to use DSPy, together with MLflow. This is built in partnership with Databricks and taught by Chen Qian, co-lead of the DSPy framework. Many AI builders spend hours hand-tuning prompts. When given a set of evals, DSPy automates this process. It’s especially useful for optimizing prompts, including few-shot prompts, in complex agentic AI workflows. Further, if you switch an application to a newer LLM, performance can degrade if your prompts were optimized to the previous model. DSPy automatically optimizes the entire system for the new LLM as well, using just a few evaluation examples. This course teaches DSPy works, and best practices for using it. You’ll write programs using DSPy’s signature-based programming model, debug them with MLflow tracing -- to gain visibility into how different parts of a pipeline, as well as how the overall system, are performing -- and automatically improve their accuracy with DSPy Optimizer. Please sign up here:

Andrew Ng

181,457 Aufrufe • vor 1 Jahr

How can you solve complex tasks using a Large Language Model? Here is a 2-minute introduction to everything you need to know to 10x the quality of your results. Let's talk about three techniques, in order of complexity, starting with the easiest one: • In-Context Learning • Indexing + In-Context Learning • Fine-tuning In-Context Learning The team that trained GPT-3 found something they couldn't explain: You can condition a model using examples of how you want it to behave. I included an example prompt in the attached video. You can "teach" the model how you want it to interpret questions, select the correct answers, and format the results by giving a few examples. You can also give specific knowledge to the model that will be helpful when formulating answers. We call this approach "grounding the model." There's another example in the video. Indexing + In-Context Learning Unfortunately, there is a limit to how much data you can include in a prompt. We call this the "context size." One version of GPT-4 supports a context of approximately 6,000 words, while the other supports 25,000 words. Although this sounds like a lot, many applications need more than that. Imagine you wrote a book and want to build an application to answer any questions about your story. What happens if your book is longer than the context? That's where Indexing comes in. Using a model, you can turn every book passage into an embedding. These are vectors, numbers that "encode" the passage's text. You can then store these embeddings in a particular database that supports fast retrieval of these vectors. You can then turn any question into an embedding and search the database for the list of passages that are similar to that query. Instead of using the entire book to ask the model, you can now use the relevant passages as in-context information, effectively working around the context size limitation. Fine-tuning Fine-tuning can give you an extra boost to get reliable outputs from your LLM. It is, however, the most complex approach on the list. There are different approaches to fine-tuning a model with your data. A popular technique is to process your data with your LLM and use the outputs to train a new classifier that solves your specific task. Notice that here you aren't modifying the LLM. Instead, you are chaining it with your trained classifier. Another approach is to modify the parameters of the LLM using your data. Think of this as "rewiring" the model in a way that solves your particular task. The results and costs will vary depending on how many layers you want to fine-tune from the original model. Many companies think that fine-tuning is the solution to their problems. In my experience, many will benefit from exploring the other two approaches. I love explaining Machine Learning and Artificial Intelligence ideas. If you enjoy in-depth content like this, follow me Santiago so you don't miss what comes next.

How can you solve complex tasks using a Large Language Model? Here is a 2-minute introduction to everything you need to know to 10x the quality of your results. Let's talk about three techniques, in order of complexity, starting with the easiest one: • In-Context Learning • Indexing + In-Context Learning • Fine-tuning In-Context Learning The team that trained GPT-3 found something they couldn't explain: You can condition a model using examples of how you want it to behave. I included an example prompt in the attached video. You can "teach" the model how you want it to interpret questions, select the correct answers, and format the results by giving a few examples. You can also give specific knowledge to the model that will be helpful when formulating answers. We call this approach "grounding the model." There's another example in the video. Indexing + In-Context Learning Unfortunately, there is a limit to how much data you can include in a prompt. We call this the "context size." One version of GPT-4 supports a context of approximately 6,000 words, while the other supports 25,000 words. Although this sounds like a lot, many applications need more than that. Imagine you wrote a book and want to build an application to answer any questions about your story. What happens if your book is longer than the context? That's where Indexing comes in. Using a model, you can turn every book passage into an embedding. These are vectors, numbers that "encode" the passage's text. You can then store these embeddings in a particular database that supports fast retrieval of these vectors. You can then turn any question into an embedding and search the database for the list of passages that are similar to that query. Instead of using the entire book to ask the model, you can now use the relevant passages as in-context information, effectively working around the context size limitation. Fine-tuning Fine-tuning can give you an extra boost to get reliable outputs from your LLM. It is, however, the most complex approach on the list. There are different approaches to fine-tuning a model with your data. A popular technique is to process your data with your LLM and use the outputs to train a new classifier that solves your specific task. Notice that here you aren't modifying the LLM. Instead, you are chaining it with your trained classifier. Another approach is to modify the parameters of the LLM using your data. Think of this as "rewiring" the model in a way that solves your particular task. The results and costs will vary depending on how many layers you want to fine-tune from the original model. Many companies think that fine-tuning is the solution to their problems. In my experience, many will benefit from exploring the other two approaches. I love explaining Machine Learning and Artificial Intelligence ideas. If you enjoy in-depth content like this, follow me Santiago so you don't miss what comes next.

Santiago

384,510 Aufrufe • vor 3 Jahren

Ok, so the new Midjourney Style Tuner is here and it's a MASSIVE update. There is so much to unpack, no way I can do it all in one post, but here is an overview. You start by typing in /tune followed by your prompt. MJ will generate a range of sample images (between 16-128) showing different visual styles based on your prompt. Then you choose your favorite images from the sample to generate a unique code you can use to customize the look of future jobs. (link below) So for example, I just tried it with the prompt: 💬 1990s, a photo of a woman dressed in fuzzy velour, shimmering midnight hues reflecting sadness by the lake, a contemporary nightmare on cross-processed film --ar 4:5 --c 4 I selected 10 images, which generated the code zlPhgLCHDXol64kPlwSUzu97. I can then add it to my prompt using --style {code} But it makes more sense if you try it yourself, so try playing around with my Style Tuner for that prompt and generate your own code from your selections: Some quick notes: -You can control the level of style applied by adding the --s parameter to your prompt. -You can combine multiple styles together by separating with a hyphen, i.e --style code1-code2 I will be obsessing over this all day tomorrow and sharing everything I learn along the way. This is an incredible update.

Ok, so the new Midjourney Style Tuner is here and it's a MASSIVE update. There is so much to unpack, no way I can do it all in one post, but here is an overview. You start by typing in /tune followed by your prompt. MJ will generate a range of sample images (between 16-128) showing different visual styles based on your prompt. Then you choose your favorite images from the sample to generate a unique code you can use to customize the look of future jobs. (link below) So for example, I just tried it with the prompt: 💬 1990s, a photo of a woman dressed in fuzzy velour, shimmering midnight hues reflecting sadness by the lake, a contemporary nightmare on cross-processed film --ar 4:5 --c 4 I selected 10 images, which generated the code zlPhgLCHDXol64kPlwSUzu97. I can then add it to my prompt using --style {code} But it makes more sense if you try it yourself, so try playing around with my Style Tuner for that prompt and generate your own code from your selections: Some quick notes: -You can control the level of style applied by adding the --s parameter to your prompt. -You can combine multiple styles together by separating with a hyphen, i.e --style code1-code2 I will be obsessing over this all day tomorrow and sharing everything I learn along the way. This is an incredible update.

Nick St. Pierre

377,975 Aufrufe • vor 2 Jahren

This is one of the wildest things I have built with AI so far 😱 Asked Claude Fable 5 to create a fully interactive 3D crime investigation board using Three.js and it did it in ONE Prompt. It took one more prompt to give it a polished UI inspired by a reference design that I created with GPT Image. What surprised me most was not how it looked. It was how it worked. Fable built a complete Verlet rope physics system with 46 particles in every thread. It handled gravity, damping, distance constraints and all other features of file upload, text and everything. The threads/ropes even lift naturally with pinned evidence instead of clipping through it. I never asked it to implement any of that. I only described the experience I wanted. Maybe that's how we should work with AI. Instead of writing long long specs, we should explain what we want as an experience and let AI do its thing. Anyway, every day Fable 5 raises the bar for what I think AI can build. More to this Live: Code:

This is one of the wildest things I have built with AI so far 😱 Asked Claude Fable 5 to create a fully interactive 3D crime investigation board using Three.js and it did it in ONE Prompt. It took one more prompt to give it a polished UI inspired by a reference design that I created with GPT Image. What surprised me most was not how it looked. It was how it worked. Fable built a complete Verlet rope physics system with 46 particles in every thread. It handled gravity, damping, distance constraints and all other features of file upload, text and everything. The threads/ropes even lift naturally with pinned evidence instead of clipping through it. I never asked it to implement any of that. I only described the experience I wanted. Maybe that's how we should work with AI. Instead of writing long long specs, we should explain what we want as an experience and let AI do its thing. Anyway, every day Fable 5 raises the bar for what I think AI can build. More to this Live: Code:

The Bugged Dev

102,570 Aufrufe • vor 23 Tagen

I built an agent that answers machine-learning questions. It's autonomous, and the best part is that I built the whole thing without writing a single line of Python code. Here is what I did and how I did it: Over a year ago, a friend and I built a site that publishes multi-choice questions. You get a new one every day. I decided to have GPT-3.5 answer questions. Here is what I needed to build: 1. Connect to the site's API to retrieve today's question 2. Extract the question and the potential choices 3. Connect to OpenAI's API and ask GPT-3.5 to answer the question 4. Parse the answer from the model 5. Submit the answer back to the API to get the score Not difficult. Likely several hours of work. But I didn't have to write any code. I built the whole thing by dragging and dropping components using Vellum is a YC-backed platform for developers to build LLM applications. They are the only ones I've seen offering this functionality. They sponsored this post, and their team helped me with all my questions while I built this. I created a workflow. The platform supports several node types to build whatever you have in mind. I show how I put the whole thing together in the attached video. The only code I had to write was a few lines of Jinja to parse and transform the API and the LLM results. There are three lessons I want to share from this experience: First, the best possible code is the one you didn't write. I'm a big fan of no-code tools because they help me materialize my ideas fast. They help product people, designers, and no coders collaborate on the solution. Second, Large Language Models are sensitive to how you prompt them. Small changes to prompts can make a big difference in results. This is more pronounced when you are building a multi-step workflow. Third, automated testing and evaluation for prompts is critical. There aren't many companies thinking about this. They'll have a hard time moving from a demo phase. The attached video will show you what I did.

I built an agent that answers machine-learning questions. It's autonomous, and the best part is that I built the whole thing without writing a single line of Python code. Here is what I did and how I did it: Over a year ago, a friend and I built a site that publishes multi-choice questions. You get a new one every day. I decided to have GPT-3.5 answer questions. Here is what I needed to build: 1. Connect to the site's API to retrieve today's question 2. Extract the question and the potential choices 3. Connect to OpenAI's API and ask GPT-3.5 to answer the question 4. Parse the answer from the model 5. Submit the answer back to the API to get the score Not difficult. Likely several hours of work. But I didn't have to write any code. I built the whole thing by dragging and dropping components using Vellum is a YC-backed platform for developers to build LLM applications. They are the only ones I've seen offering this functionality. They sponsored this post, and their team helped me with all my questions while I built this. I created a workflow. The platform supports several node types to build whatever you have in mind. I show how I put the whole thing together in the attached video. The only code I had to write was a few lines of Jinja to parse and transform the API and the LLM results. There are three lessons I want to share from this experience: First, the best possible code is the one you didn't write. I'm a big fan of no-code tools because they help me materialize my ideas fast. They help product people, designers, and no coders collaborate on the solution. Second, Large Language Models are sensitive to how you prompt them. Small changes to prompts can make a big difference in results. This is more pronounced when you are building a multi-step workflow. Third, automated testing and evaluation for prompts is critical. There aren't many companies thinking about this. They'll have a hard time moving from a demo phase. The attached video will show you what I did.

Santiago

309,825 Aufrufe • vor 2 Jahren

SITUATION EXPLAINED: Why can't Anthropic just build Claude for Law and cut out law firms entirely? We asked Zack Shapiro, managing partner at Rains "I own a law practice. I use AI in every part of my law practice. I can pretty confidently say that the best Claude for Law is just Claude." "Claude is really, really excellent at law. You don't need a different harness. You don't need different post-training." "The thing that makes for good performance with Claude on legal tasks is actually the input rather than the output it's trained on. It is the prompt you give it. It is the instructions. It's the custom skills and workflows that you really do need a lawyer to build as part of their own practice." "I used to spend four to six hours a day in front of Microsoft Word, hunched over my desk. Now when I do my most difficult cognitive work, I'll go outside, go for a walk, free associate into the microphone for three to four minutes. By the time I get back to my desk, I'll have a first draft."

SITUATION EXPLAINED: Why can't Anthropic just build Claude for Law and cut out law firms entirely? We asked Zack Shapiro, managing partner at Rains "I own a law practice. I use AI in every part of my law practice. I can pretty confidently say that the best Claude for Law is just Claude." "Claude is really, really excellent at law. You don't need a different harness. You don't need different post-training." "The thing that makes for good performance with Claude on legal tasks is actually the input rather than the output it's trained on. It is the prompt you give it. It is the instructions. It's the custom skills and workflows that you really do need a lawyer to build as part of their own practice." "I used to spend four to six hours a day in front of Microsoft Word, hunched over my desk. Now when I do my most difficult cognitive work, I'll go outside, go for a walk, free associate into the microphone for three to four minutes. By the time I get back to my desk, I'll have a first draft."

MTS

143,964 Aufrufe • vor 1 Monat

This is how you get 15x fable 5 usage. Fable -> Composer 2.5 reading Fable -> GPT 5.5 execution I built CNVS to make agent orchestration dead simple and insanely visual. You can watch fable 5 delegate work to cursor, codex, open code, all your agents. using your exsisting ai subs no api pricing. The canvas is fully voice controlled locally with nvidia parakeet for fast and free, or gpt realtime 2 for a fully conversational jarvis experience. The built in mcp and cli means agentic control is bidirectional they can prompt and spawn each other and read agent states. but there is more I built it from the ground up in swift for native performance on mac os. My old 16gb m1 mbp can easily spin up and control eight agents across multiple canvasses. The hermes integration + remote canvases allow you to run agents fully in the cloud on your VPS think google docs of vibe coding. You can literally turn off your mac and they keep working, open cnvs back up and pick up where you left off. I built a cross agent memory system based on 2026 research so all your agents feel like one brain. its SIMPLE and on demand. This eliminates context bloat, and the research supports a massive uptick in cross agent performance. There is soo much more. PS - I am a father building this project 100% solo in my basement on live stream everyday, my goal is to push vibe coding to the next level PPS - CNVS is a lifetime license becuase you DO NOT need another subscription in your life.

This is how you get 15x fable 5 usage. Fable -> Composer 2.5 reading Fable -> GPT 5.5 execution I built CNVS to make agent orchestration dead simple and insanely visual. You can watch fable 5 delegate work to cursor, codex, open code, all your agents. using your exsisting ai subs no api pricing. The canvas is fully voice controlled locally with nvidia parakeet for fast and free, or gpt realtime 2 for a fully conversational jarvis experience. The built in mcp and cli means agentic control is bidirectional they can prompt and spawn each other and read agent states. but there is more I built it from the ground up in swift for native performance on mac os. My old 16gb m1 mbp can easily spin up and control eight agents across multiple canvasses. The hermes integration + remote canvases allow you to run agents fully in the cloud on your VPS think google docs of vibe coding. You can literally turn off your mac and they keep working, open cnvs back up and pick up where you left off. I built a cross agent memory system based on 2026 research so all your agents feel like one brain. its SIMPLE and on demand. This eliminates context bloat, and the research supports a massive uptick in cross agent performance. There is soo much more. PS - I am a father building this project 100% solo in my basement on live stream everyday, my goal is to push vibe coding to the next level PPS - CNVS is a lifetime license becuase you DO NOT need another subscription in your life.

Max Blade

29,504 Aufrufe • vor 29 Tagen

someone just rebuilt claude code in pure python and made it ~230x cheaper to run it's called clawcodex. a full from-scratch port of the claude code agent loop. 230k+ lines of python, MIT licensed it keeps your request prefix byte-stable, so DeepSeek's prompt cache covers your entire system + tools + history on every turn cache-hit input then bills at ~$0.0435 per 1M tokens the same 1M tokens on Claude Fable 5 runs you $10 so the longer you code, the more the cache pays off what you actually get: → 25 LLM providers (Claude, GPT, DeepSeek, GLM, local ollama) → 1M token context window → real agent loop: tools, skills, REPL, session history → one-line install, mac/linux/wsl and it's sitting at 677 stars. basically nobody is looking yet bookmark this before it gone

someone just rebuilt claude code in pure python and made it ~230x cheaper to run it's called clawcodex. a full from-scratch port of the claude code agent loop. 230k+ lines of python, MIT licensed it keeps your request prefix byte-stable, so DeepSeek's prompt cache covers your entire system + tools + history on every turn cache-hit input then bills at ~$0.0435 per 1M tokens the same 1M tokens on Claude Fable 5 runs you $10 so the longer you code, the more the cache pays off what you actually get: → 25 LLM providers (Claude, GPT, DeepSeek, GLM, local ollama) → 1M token context window → real agent loop: tools, skills, REPL, session history → one-line install, mac/linux/wsl and it's sitting at 677 stars. basically nobody is looking yet bookmark this before it gone

Argona

51,369 Aufrufe • vor 27 Tagen

We’re open sourcing the first document OCR benchmark for the agentic era, ParseBench. Document parsing is the foundation of every AI agent that works with real-world files. ParseBench is a benchmark that measures parsing quality specifically for agent knowledge work: ✅ It optimizes for semantic correctness (instead of exact similarity) ✅ It has the most comprehensive distribution of real-world enterprise documents It contains ~2,000 human-verified enterprise document pages with 167,000+ test rules across five dimensions that matter most: tables, charts, content faithfulness, semantic formatting, and visual grounding. We benchmarked 14 known document parsers on ParseBench, from frontier/OSS VLMs to specialized parsers to LlamaParse. Here are some of our findings: 💡 Increasing compute budget yields diminishing returns - Gemini/gpt-5-mini/haiku gain 3-5 points from minimal to high thinking, at 4x the cost. 💡 Charts are the most polarizing dimension for evaluation. Most specialized parsers score below 6%, while some VLM-based parsers do a bit better. 💡 VLMs are great at visual understanding but terrible at layout extraction. GPT-5-mini/haiku score below 10% on our visual grounding task, all specialized parsers do much better. 💡 No method crushes all 5 dimensions at once, but LlamaParse achieves the highest overall score at 84.9%, and is the leader in 4 out of the 5 dimensions. This is by far the deepest technical work that we’ve published as a company. I would encourage you to start with our blog and explore our links to Hugging Face to GitHub. All the details are in our full 35-page (!!) ArXiv whitepaper. 🌐: Blog: 📄 Paper: 💻 Code: 📊 Dataset: 🎥 YouTube:

We’re open sourcing the first document OCR benchmark for the agentic era, ParseBench. Document parsing is the foundation of every AI agent that works with real-world files. ParseBench is a benchmark that measures parsing quality specifically for agent knowledge work: ✅ It optimizes for semantic correctness (instead of exact similarity) ✅ It has the most comprehensive distribution of real-world enterprise documents It contains ~2,000 human-verified enterprise document pages with 167,000+ test rules across five dimensions that matter most: tables, charts, content faithfulness, semantic formatting, and visual grounding. We benchmarked 14 known document parsers on ParseBench, from frontier/OSS VLMs to specialized parsers to LlamaParse. Here are some of our findings: 💡 Increasing compute budget yields diminishing returns - Gemini/gpt-5-mini/haiku gain 3-5 points from minimal to high thinking, at 4x the cost. 💡 Charts are the most polarizing dimension for evaluation. Most specialized parsers score below 6%, while some VLM-based parsers do a bit better. 💡 VLMs are great at visual understanding but terrible at layout extraction. GPT-5-mini/haiku score below 10% on our visual grounding task, all specialized parsers do much better. 💡 No method crushes all 5 dimensions at once, but LlamaParse achieves the highest overall score at 84.9%, and is the leader in 4 out of the 5 dimensions. This is by far the deepest technical work that we’ve published as a company. I would encourage you to start with our blog and explore our links to Hugging Face to GitHub. All the details are in our full 35-page (!!) ArXiv whitepaper. 🌐: Blog: 📄 Paper: 💻 Code: 📊 Dataset: 🎥 YouTube:

Jerry Liu

108,011 Aufrufe • vor 3 Monaten

Fable 5 and GPT-5.6 built the same scroll-animated website from 1 skill in 32 minutes and only 1 of them made it feel like a film. Same prompt: boutique Japan travel brand, origami style. A subway pulls in, a paper house unfolds into a hotel, a bird takes flight as you scroll. Doing this by hand is brutal multiple videos, matched starting frames, frames ripped out 1 by 1 and synced to scroll position. The skill does all of it: Generate 1 anchor image and approve it every scene inherits the style Turn it into video through the Higgs Field MCP (Seedance), straight from the terminal FFmpeg pulls every frame Each frame maps to scroll position, so your scrollbar becomes the playhead Setup is just connecting the MCP and loading the skill works in Claude Code and Codex. It interviews you first about scenes, budget and mobile, then shows the anchor image before burning a single credit on video. Budget reality: 6 scenes runs ~800 credits and is overkill, 4 scenes is the sweet spot, crop-safe mobile is the cheap path. The verdict came down to transitions. GPT-5.6 Soul built strong scenes with incredible detail inside each one then hard cuts between scene 1 and 2, and again between 2 and 3. Fable 5 stitched them together: wires push out of the top of frame while the next scene blurs in behind, gains depth and locks into place. Same skill, same prompts, 10 generations each. Credit to Peter Wang for the original Scroll World skill open source, now forked with budget tiers and mobile fixes. Soul built 4 scenes. Fable built 1 film.

Spike 1%

44,834 Aufrufe • vor 21 Tagen

Dylan Patel of SemiAnalysis says a worse GPU with better storage and memory now beats the best chip without them, so buying the newest GPU alone no longer wins inference. So, an AMD GPU with more memory can outperform Nvidia in some cases. "So what we have is we have over $80 million of compute, GPUs from Nvidia, AMD, TPUs from Google, Trainium from Amazon, and we run this benchmark constantly on the newest inference engine, newest drivers, newest PyTorch version, whatever it is." "Every day it runs on an automated CI, and we run it on all the latest Chinese models, from GLM, Zhipu, Moonshot, Kimi, Alibaba, all these models we run." "Initially, when we were benchmarking the difference between these chips and different engines, different schemes for parallelism, we were just running it fixed context length." "But now with Agent X, we've analyzed over $5 million worth of Claude Code traces. This is real production traffic that people have donated to us as well as internally generated. Now we know what the actual agent workload looks like." "And then as we implement that and run those benchmarks, it turns out yes, the chip you're using is very important, but now even more important is how are you handling this memory offload?" "And so while an Nvidia GPU is faster than an AMD GPU in most cases, because AMD GPUs have more memory, they actually end up outperforming in some cases." "Or you can have a worse GPU, but a much better storage solution, and now you can outperform what the best GPU can do without those solutions. So just buying the newest and latest GPU alone doesn't get you the best inference economics." "Actually, you need to layer in all these other innovations including storage and memory." [ Who's the top player on your chart? ] "That really is a difficult multivariable problem. And generally that means you need to have, yes, you need to have the best GPU, a GB300, but you also need to have the best storage solutions. And so I won't spoil who's the best right here, but I will say that storage solutions matter a lot and memory solutions matter a lot, as does your front-end networking. That matters a lot."

Dylan Patel of SemiAnalysis says a worse GPU with better storage and memory now beats the best chip without them, so buying the newest GPU alone no longer wins inference. So, an AMD GPU with more memory can outperform Nvidia in some cases. "So what we have is we have over $80 million of compute, GPUs from Nvidia, AMD, TPUs from Google, Trainium from Amazon, and we run this benchmark constantly on the newest inference engine, newest drivers, newest PyTorch version, whatever it is." "Every day it runs on an automated CI, and we run it on all the latest Chinese models, from GLM, Zhipu, Moonshot, Kimi, Alibaba, all these models we run." "Initially, when we were benchmarking the difference between these chips and different engines, different schemes for parallelism, we were just running it fixed context length." "But now with Agent X, we've analyzed over $5 million worth of Claude Code traces. This is real production traffic that people have donated to us as well as internally generated. Now we know what the actual agent workload looks like." "And then as we implement that and run those benchmarks, it turns out yes, the chip you're using is very important, but now even more important is how are you handling this memory offload?" "And so while an Nvidia GPU is faster than an AMD GPU in most cases, because AMD GPUs have more memory, they actually end up outperforming in some cases." "Or you can have a worse GPU, but a much better storage solution, and now you can outperform what the best GPU can do without those solutions. So just buying the newest and latest GPU alone doesn't get you the best inference economics." "Actually, you need to layer in all these other innovations including storage and memory." [ Who's the top player on your chart? ] "That really is a difficult multivariable problem. And generally that means you need to have, yes, you need to have the best GPU, a GB300, but you also need to have the best storage solutions. And so I won't spoil who's the best right here, but I will say that storage solutions matter a lot and memory solutions matter a lot, as does your front-end networking. That matters a lot."

Fireside Alpha

170,760 Aufrufe • vor 7 Tagen

google flow + claude + gpt image is actually f*cking cracked i figured out how to lock one AI character across an entire video. same face, same voice, same outfit, every single shot. zero drift. everyone making AI video right now hits the same wall. the character looks different in every clip so you can never tell an actual story. you get 5 second clips, not films. the fix isn't better prompting. it's building the character ONCE before you ever touch video, then referencing it everywhere. here's the workflow: > build the character inside Flow's character section, not a per-clip prompt > design the voice + personality one time and reuse it across every clip > generate a full turnaround sheet (Claude writes the prompt, GPT image renders the angles) > pull the character into any scene with a single @ tag > ingredients vs frames: when to let a shot float vs lock it to your source image > the end-frame loop that chains 4-10 sec clips into one seamless scene > stitch + trim in the timeline editor and export the whole thing as one film i packaged the entire thing. every prompt, the exact tool order, the character sheet template, and the continuity loop nobody actually explains. RT + reply "FLOW" and i'll send it over (must be following so i can dm)

google flow + claude + gpt image is actually f*cking cracked i figured out how to lock one AI character across an entire video. same face, same voice, same outfit, every single shot. zero drift. everyone making AI video right now hits the same wall. the character looks different in every clip so you can never tell an actual story. you get 5 second clips, not films. the fix isn't better prompting. it's building the character ONCE before you ever touch video, then referencing it everywhere. here's the workflow: > build the character inside Flow's character section, not a per-clip prompt > design the voice + personality one time and reuse it across every clip > generate a full turnaround sheet (Claude writes the prompt, GPT image renders the angles) > pull the character into any scene with a single @ tag > ingredients vs frames: when to let a shot float vs lock it to your source image > the end-frame loop that chains 4-10 sec clips into one seamless scene > stitch + trim in the timeline editor and export the whole thing as one film i packaged the entire thing. every prompt, the exact tool order, the character sheet template, and the continuity loop nobody actually explains. RT + reply "FLOW" and i'll send it over (must be following so i can dm)

Sulfur

35,357 Aufrufe • vor 1 Monat

THIS GUY BUILT AN AUTONOMOUS AI AGENT OUT OF CLAUDE CODE + OBSIDIAN and this is way more interesting than another “use AI to take notes” demo the trick is simple: Obsidian is not the writing app here. it becomes the agent’s memory, task board, and context folder. Claude Code is not just answering prompts. it reads the vault, edits files, follows instructions, and keeps moving through the work like a junior operator with a filesystem. the reusable setup looks like this: 1. create an Obsidian vault for one project 2. keep goals, rules, tasks, decisions, and references as markdown files 3. point Claude Code at the folder 4. give it a clear operating loop: read context → choose next task → execute → write back what changed 5. use the notes as persistent memory instead of re-explaining the project every chat that’s the part people miss. the “agent” is not magic. it’s the boring combination of: - local files - explicit rules - task state - write access - a model that can run through the repo/vault Obsidian makes the memory human-readable. Claude Code makes the memory executable. that combo is why the video worked: it turns a notes app into an operating surface for actual work. best use cases: - content systems - research vaults - coding projects - client ops docs - personal knowledge bases that need actions, not just storage the caveat: if your vault is messy, your agent becomes messy too. folders, naming, “done” criteria, and forbidden actions matter more than the prompt. but once the structure is clean, this is one of the easiest ways to build an agent that remembers what happened yesterday without paying for a full custom app.

THIS GUY BUILT AN AUTONOMOUS AI AGENT OUT OF CLAUDE CODE + OBSIDIAN and this is way more interesting than another “use AI to take notes” demo the trick is simple: Obsidian is not the writing app here. it becomes the agent’s memory, task board, and context folder. Claude Code is not just answering prompts. it reads the vault, edits files, follows instructions, and keeps moving through the work like a junior operator with a filesystem. the reusable setup looks like this: 1. create an Obsidian vault for one project 2. keep goals, rules, tasks, decisions, and references as markdown files 3. point Claude Code at the folder 4. give it a clear operating loop: read context → choose next task → execute → write back what changed 5. use the notes as persistent memory instead of re-explaining the project every chat that’s the part people miss. the “agent” is not magic. it’s the boring combination of: - local files - explicit rules - task state - write access - a model that can run through the repo/vault Obsidian makes the memory human-readable. Claude Code makes the memory executable. that combo is why the video worked: it turns a notes app into an operating surface for actual work. best use cases: - content systems - research vaults - coding projects - client ops docs - personal knowledge bases that need actions, not just storage the caveat: if your vault is messy, your agent becomes messy too. folders, naming, “done” criteria, and forbidden actions matter more than the prompt. but once the structure is clean, this is one of the easiest ways to build an agent that remembers what happened yesterday without paying for a full custom app.

kocer

30,403 Aufrufe • vor 1 Monat