Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

You don’t have to use one model (or one provider!) for everything. With Workshop, you can combine frontier and local models in the same workflow. For example: Opus can be the main agent, and delegate specific tasks to Gemma 4 via subagents. Better quality where it matters. Better privacy,... show more

Workshop AI

2,340 subscribers

962,487 views • 5 months ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

subagents are just recursive agents where you can apply different prompts + models depending on the task. since they’re just a primitive, Cursor cli can actually spawn subagents by calling cursor-agent in headless mode via shell commands. that’s what makes the cli so nice. you can extend it, experiment, and have a lot of fun exploring orchestration patterns. here’s one way to do it w. dynamic model selection: 1. create a subagents.mdc rule 2. drop in: ``` --- alwaysApply: true --- ALWAYS spawn subagents by running `cursor-agent -p [task] --output-format=text --force --model [model]` in the terminal. Each subagent should return a summary of the changes it made. Subagents should be used for ALL tasks You can adopt a fan-out pattern where you spawn subagents to perform parallel isolated tasks, and then fan-in the results. Use the following models: - `--model gpt-5` for reasoning, researching, and planning - `--model sonnet-4` for implementation ``` 3. start cursor cli and try it out you can also adjust the rule to be more explicit when it should use subagents, when not to, which models when etc.

subagents are just recursive agents where you can apply different prompts + models depending on the task. since they’re just a primitive, Cursor cli can actually spawn subagents by calling cursor-agent in headless mode via shell commands. that’s what makes the cli so nice. you can extend it, experiment, and have a lot of fun exploring orchestration patterns. here’s one way to do it w. dynamic model selection: 1. create a subagents.mdc rule 2. drop in: ``` --- alwaysApply: true --- ALWAYS spawn subagents by running `cursor-agent -p [task] --output-format=text --force --model [model]` in the terminal. Each subagent should return a summary of the changes it made. Subagents should be used for ALL tasks You can adopt a fan-out pattern where you spawn subagents to perform parallel isolated tasks, and then fan-in the results. Use the following models: - `--model gpt-5` for reasoning, researching, and planning - `--model sonnet-4` for implementation ``` 3. start cursor cli and try it out you can also adjust the rule to be more explicit when it should use subagents, when not to, which models when etc.

eric zakariasson

57,554 views • 1 year ago

Devin for Terminal is a local agent that works with all frontier models, including Opus 4.7, GPT 5.5, and SWE-1.6. You can switch model mid-session, or handoff to Devin in the cloud.

Devin for Terminal is a local agent that works with all frontier models, including Opus 4.7, GPT 5.5, and SWE-1.6. You can switch model mid-session, or handoff to Devin in the cloud.

Cognition

19,795,575 views • 4 months ago

Not every task needs the same model. A quick summary doesn't need the same horsepower as a deep research question — and it shouldn't cost the same either. Now, you can now easily compare models for your Custom Agent on speed, intelligence, and cost 🫡

Not every task needs the same model. A quick summary doesn't need the same horsepower as a deep research question — and it shouldn't cost the same either. Now, you can now easily compare models for your Custom Agent on speed, intelligence, and cost 🫡

Notion

32,420 views • 6 months ago

MiniMax H3 vs Seedance 2.0 Created entirely on Pollo AI using the same prompt and settings to compare both models fairly. From commercial visuals to camera movement, each model brings a different style. Watch the comparison and decide which one performs better for your workflow. Which one would you choose—MiniMax H3 or Seedance 2.0?

MiniMax H3 vs Seedance 2.0 Created entirely on Pollo AI using the same prompt and settings to compare both models fairly. From commercial visuals to camera movement, each model brings a different style. Watch the comparison and decide which one performs better for your workflow. Which one would you choose—MiniMax H3 or Seedance 2.0?

Sophia

41,157 views • 1 month ago

Multi-bot chat allows you to seamlessly combine the unique strengths of different bots to complete tasks. For example, using fal’s latest bots, you can generate an image and remove the background within one conversational thread for a more efficient workflow. (1/2)

Multi-bot chat allows you to seamlessly combine the unique strengths of different bots to complete tasks. For example, using fal’s latest bots, you can generate an image and remove the background within one conversational thread for a more efficient workflow. (1/2)

Poe

19,538 views • 2 years ago

I built a GPT 2 workflow that turns one image into 20 viral ads GPT image 2 is the most powerful image model out there And you can use it to scale your ads campaign with Arcads workflow Comment Arcads to get the workflow now

I built a GPT 2 workflow that turns one image into 20 viral ads GPT image 2 is the most powerful image model out there And you can use it to scale your ads campaign with Arcads workflow Comment Arcads to get the workflow now

Romain Torres

80,386 views • 5 months ago

New open-source agent harness just landed! I got early access to TrueForge by TrueFoundry and have been running it locally for the past few days. The harness layer deserves as much attention as the model, and open source matters here because you can inspect the loop, run it on your own infrastructure, and swap to the latest or cheaper models. TrueForge handles the runtime work that makes an agent reliable. It drives the tool-calling loop, manages context, coordinates subagents, and executes code in a sandbox, with any model you choose. Every tool call re-sends the growing context to the model, so in practice the harness controls most of what an agent costs to run. A few things stood out from my testing and their published benchmarks. Vendor-Neutral by design. It runs OpenAI, Anthropic, and Google models alongside open-weight models like Kimi, GLM, and DeepSeek. Model routing is a setting, and you can send each task to the model that fits it. On a 14-task enterprise agent benchmark, it matched the accuracy of Claude Managed Agents running the same Opus 4.8 model at roughly 30% lower cost per run (3.8M tokens vs 10M for the same answers). Routing the same tasks to GLM-5.2 held accuracy and brought cost down by about 75%, around $3 per run instead of $12. Fully self-hosted and Open Source (MIT License). I had it running locally with one command, with sandboxed code execution working out of the box. It's time to own your agent harness. Thanks to TrueFoundry for partnering on this post.

New open-source agent harness just landed! I got early access to TrueForge by TrueFoundry and have been running it locally for the past few days. The harness layer deserves as much attention as the model, and open source matters here because you can inspect the loop, run it on your own infrastructure, and swap to the latest or cheaper models. TrueForge handles the runtime work that makes an agent reliable. It drives the tool-calling loop, manages context, coordinates subagents, and executes code in a sandbox, with any model you choose. Every tool call re-sends the growing context to the model, so in practice the harness controls most of what an agent costs to run. A few things stood out from my testing and their published benchmarks. Vendor-Neutral by design. It runs OpenAI, Anthropic, and Google models alongside open-weight models like Kimi, GLM, and DeepSeek. Model routing is a setting, and you can send each task to the model that fits it. On a 14-task enterprise agent benchmark, it matched the accuracy of Claude Managed Agents running the same Opus 4.8 model at roughly 30% lower cost per run (3.8M tokens vs 10M for the same answers). Routing the same tasks to GLM-5.2 held accuracy and brought cost down by about 75%, around $3 per run instead of $12. Fully self-hosted and Open Source (MIT License). I had it running locally with one command, with sandboxed code execution working out of the box. It's time to own your agent harness. Thanks to TrueFoundry for partnering on this post.

elvis

11,303 views • 1 month ago

Perplexity Computer is massively multi-model. Computer orchestrates models to run agents in parallel, leveraging Opus to match each task to the model best suited for it. In total, Computer can route work across 19 different models.

Perplexity Computer is massively multi-model. Computer orchestrates models to run agents in parallel, leveraging Opus to match each task to the model best suited for it. In total, Computer can route work across 19 different models.

Perplexity

534,569 views • 6 months ago

Quality, speed, efficiency. Now you get all three, without having to choose. Intelligent Model Routing picks the best model for every task, so you don’t have to. That means top tier output for up to 65% less cost. 0 effort spent switching models.

Quality, speed, efficiency. Now you get all three, without having to choose. Intelligent Model Routing picks the best model for every task, so you don’t have to. That means top tier output for up to 65% less cost. 0 effort spent switching models.

Replit ⠕

13,822 views • 26 days ago

Tasks can now spawn async subagents that move to the background and continue working independently, even if the main agent finishes its task and becomes inactive. This lets subagents handle long-running tasks, great for monitoring logs or waiting for builds.

Tasks can now spawn async subagents that move to the background and continue working independently, even if the main agent finishes its task and becomes inactive. This lets subagents handle long-running tasks, great for monitoring logs or waiting for builds.

Claude

349,678 views • 9 months ago

Open weight models are now available in Kiro. Three open weight models are available in the IDE and CLI, so you don’t have to default to the largest model for every task. Choose the right model based on speed, cost, or long-running agentic workflows. What model would you like to see supported next? #Kiro #BuildwithKiro #AICoding

Open weight models are now available in Kiro. Three open weight models are available in the IDE and CLI, so you don’t have to default to the largest model for every task. Choose the right model based on speed, cost, or long-running agentic workflows. What model would you like to see supported next? #Kiro #BuildwithKiro #AICoding

Kiro

22,214 views • 7 months ago

Hermes Desktop now sets up local models in one click. It automatically reads your hardware, picks the best model for you, then downloads it and configures the runtime.

Hermes Desktop now sets up local models in one click. It automatically reads your hardware, picks the best model for you, then downloads it and configures the runtime.

Nous Research

411,112 views • 19 days ago

OpenClaw can use Autobrowse to create and iteratively improve a Skill for any workflow. In this Craigslist extraction example we found a 68% speed up and 91% cost savings in 5 iterations. Halfway through, the agent found an exposed endpoint to speed up page navigation.

OpenClaw can use Autobrowse to create and iteratively improve a Skill for any workflow. In this Craigslist extraction example we found a 68% speed up and 91% cost savings in 5 iterations. Halfway through, the agent found an exposed endpoint to speed up page navigation.

Kyle Jeong

108,724 views • 4 months ago

Fusion is the most efficient frontier harness for GPT 6 Astra and Claude Fable 5.1. You can also configure everything about it: • Base model: Fable, Astra, Sol, or Opus • Sidekick model: SWE (Free), Luna, Sol, GLM (Free) • Speed: Normal or Fast ⚡ • Reasoning level It retains Claude Fable 5.1 and GPT-6 Astra performance while reducing costs and is the first time a multi-model coding agent has been included on the Artificial Analysis Coding Agent Index! Importantly: it's also fun to use :)

Fusion is the most efficient frontier harness for GPT 6 Astra and Claude Fable 5.1. You can also configure everything about it: • Base model: Fable, Astra, Sol, or Opus • Sidekick model: SWE (Free), Luna, Sol, GLM (Free) • Speed: Normal or Fast ⚡ • Reasoning level It retains Claude Fable 5.1 and GPT-6 Astra performance while reducing costs and is the first time a multi-model coding agent has been included on the Artificial Analysis Coding Agent Index! Importantly: it's also fun to use :)

nader dabit

804,446 views • 11 days ago

You can now delegate tasks to GitHub Copilot coding agent from any page on GitHub 🤖 Open the new Agents panel in one click, write a simple prompt, then hit Enter. GitHub Copilot works in the background, and opens a PR for your review. No interruptions to your workflow required. ✅

You can now delegate tasks to GitHub Copilot coding agent from any page on GitHub 🤖 Open the new Agents panel in one click, write a simple prompt, then hit Enter. GitHub Copilot works in the background, and opens a PR for your review. No interruptions to your workflow required. ✅

GitHub

113,392 views • 1 year ago

Wan 3.0 is now available on Flova. To put it to the test, I gave Wan 3.0, Seedance 2.5, and MiniMax H3 the same live-action ice queen scene—and each model created a completely different frozen world. With Flova’s all-in-one Agent workflow, you can call multiple models, compare their results, and continue refining your film without switching platforms. Which version would you choose? FlovaAI x Wan 3.0! Limited-Time Offer for Wan3.0 480P: As Low As $0.013/sec

Wan 3.0 is now available on Flova. To put it to the test, I gave Wan 3.0, Seedance 2.5, and MiniMax H3 the same live-action ice queen scene—and each model created a completely different frozen world. With Flova’s all-in-one Agent workflow, you can call multiple models, compare their results, and continue refining your film without switching platforms. Which version would you choose? FlovaAI x Wan 3.0! Limited-Time Offer for Wan3.0 480P: As Low As $0.013/sec

K

21,742 views • 28 days ago

a moonshot engineer leaked the benchmark anthropic, openai and xai all buried the same week: kimi k3 beat opus 5, gpt-5.6 and grok 4.6 at $0.94 a task. stop paying anthropic $200 a month for opus 5 and openai $200 for gpt-5.6 when kimi does the same work for $8 the leak showed kimi k3 winning 9 of 12 categories against opus 5, gpt-5.6 and grok 4.6. within 48 hours all three labs quietly pushed pricing pages and one very specific comparison chart off their sites. nobody announced anything. they just deleted, which tells you everything the four numbers they scrubbed: cost per task · $0.94 vs $1.80 -> opus 5 charges $1.80 to finish one task. gpt-5.6 $1.04. grok 4.6 $0.61. kimi k3 $0.94 and it landed 487 of 500 clean -> anthropic is billing you double for a model that lost the benchmark it paid to promote the weights · free, sitting on huggingface right now -> the entire model is a public download. pull it, keep it, run it forever, nobody can switch it off -> a model you can hold cannot be rented at $200 a month. that single fact is what three labs deleted a chart over the switch · one line of bash -> moonshot ships an anthropic-compatible endpoint. one env variable and claude code points at kimi -> same cli, same keybindings, same /model. you change a url, opus 5 never knows it lost the seat the bill · $400 down to $8 -> opus 5 max plus gpt-5.6 pro is $400 a month. kimi runs the same daily work for $8 metered -> that is a 98% cut for output that beat both of them 9 categories to 3 here is the part they will fight me on: the frontier tax died the week this leaked and all three labs know it. once the weights are public the price has a ceiling, because anyone can serve the same model. anthropic, openai and xai are charging 2025 prices on a lead that ended in a benchmark they deleted instead of answered drop your $400/mo ai stack to $8. the run above is kimi k3 finishing the task opus 5 bills $1.80 for. the full breakdown is in the article below

a moonshot engineer leaked the benchmark anthropic, openai and xai all buried the same week: kimi k3 beat opus 5, gpt-5.6 and grok 4.6 at $0.94 a task. stop paying anthropic $200 a month for opus 5 and openai $200 for gpt-5.6 when kimi does the same work for $8 the leak showed kimi k3 winning 9 of 12 categories against opus 5, gpt-5.6 and grok 4.6. within 48 hours all three labs quietly pushed pricing pages and one very specific comparison chart off their sites. nobody announced anything. they just deleted, which tells you everything the four numbers they scrubbed: cost per task · $0.94 vs $1.80 -> opus 5 charges $1.80 to finish one task. gpt-5.6 $1.04. grok 4.6 $0.61. kimi k3 $0.94 and it landed 487 of 500 clean -> anthropic is billing you double for a model that lost the benchmark it paid to promote the weights · free, sitting on huggingface right now -> the entire model is a public download. pull it, keep it, run it forever, nobody can switch it off -> a model you can hold cannot be rented at $200 a month. that single fact is what three labs deleted a chart over the switch · one line of bash -> moonshot ships an anthropic-compatible endpoint. one env variable and claude code points at kimi -> same cli, same keybindings, same /model. you change a url, opus 5 never knows it lost the seat the bill · $400 down to $8 -> opus 5 max plus gpt-5.6 pro is $400 a month. kimi runs the same daily work for $8 metered -> that is a 98% cut for output that beat both of them 9 categories to 3 here is the part they will fight me on: the frontier tax died the week this leaked and all three labs know it. once the weights are public the price has a ceiling, because anyone can serve the same model. anthropic, openai and xai are charging 2025 prices on a lead that ended in a benchmark they deleted instead of answered drop your $400/mo ai stack to $8. the run above is kimi k3 finishing the task opus 5 bills $1.80 for. the full breakdown is in the article below

starmex

32,974 views • 1 month ago

Quick update to the KeebWarrior add-ons. Made it so that the controllers now pivot in place so that you can better align them for straight wrists to forearms. Updated model included for all current orders. If you have one that doesn't pivot and want to update to it, send me a dm

Quick update to the KeebWarrior add-ons. Made it so that the controllers now pivot in place so that you can better align them for straight wrists to forearms. Updated model included for all current orders. If you have one that doesn't pivot and want to update to it, send me a dm

Mavercade

63,112 views • 1 year ago

I've nearly ditched closed source models for image generation (in creative workflows). For most tasks, the cost/speed ratio doesn't justify closed models anymore. Open models have caught up on quality and I can try 15 new images in 1-2 sec. Iteration speed matters. Sharing the few models I currently use ⬇️

I've nearly ditched closed source models for image generation (in creative workflows). For most tasks, the cost/speed ratio doesn't justify closed models anymore. Open models have caught up on quality and I can try 15 new images in 1-2 sec. Iteration speed matters. Sharing the few models I currently use ⬇️

Victor M

39,333 views • 8 months ago