正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Deterministic inference: getting the exact same output every time you run an LLM with identical inputs. Try it yourself: Powered by EigenAI

nader dabit

187,550 subscribers

29,335 次观看 • 9 个月前 •via X (Twitter)

教育科学技术

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Introducing Summaries! Now, every Exa result can come with an LLM-generated summary, or an answer to any question you specify. Try it for yourself!

Introducing Summaries! Now, every Exa result can come with an LLM-generated summary, or an answer to any question you specify. Try it for yourself!

Exa

20,419 次观看 • 2 年前

We've built the best in house observability for Stagehand 🔭 Get inference time, token usage, LLM output, and more on the new dashboard. No more black box, see everything behind the scenes. See for yourself, get started with npx create-browser-app 🤘

We've built the best in house observability for Stagehand 🔭 Get inference time, token usage, LLM output, and more on the new dashboard. No more black box, see everything behind the scenes. See for yourself, get started with npx create-browser-app 🤘

Stagehand

10,887 次观看 • 1 年前

Let's (ab)use an LLM API with excessive agency to delete Carlos' user account! (sorry Carlos!) Try it yourself here:

Let's (ab)use an LLM API with excessive agency to delete Carlos' user account! (sorry Carlos!) Try it yourself here:

Web Security Academy

12,646 次观看 • 6 个月前

This is a very clever idea to use an LLM with your SQL data. SQL + AI has been tried before, but one of the best parts of this solution is getting the same exact OpenAI's completion API. In other words: You are now talking to an LLM that knows everything about your database and you don't even notice it! I recorded a 2-min video to show you how it works. Thanks to the team for the collaboration, listening to my feedback, and helping me understand what they built. Go to to try this yourself.

This is a very clever idea to use an LLM with your SQL data. SQL + AI has been tried before, but one of the best parts of this solution is getting the same exact OpenAI's completion API. In other words: You are now talking to an LLM that knows everything about your database and you don't even notice it! I recorded a 2-min video to show you how it works. Thanks to the team for the collaboration, listening to my feedback, and helping me understand what they built. Go to to try this yourself.

Santiago

230,503 次观看 • 1 年前

Introducing the most precise calculator in the world, powered by GPT-5.2 Try it yourself

Introducing the most precise calculator in the world, powered by GPT-5.2 Try it yourself

.davιd🎀

2,299,085 次观看 • 5 个月前

Introducing the most precise calculator in the world, powered by GPT-5.2 Try it yourself

Introducing the most precise calculator in the world, powered by GPT-5.2 Try it yourself

Ann Nguyen

24,083,227 次观看 • 7 个月前

Right now, I have a few dozen tools to automate my life. Every single thing that I do more than once every week is now a tool: either a Claude Code skill, a scheduled workflow, or an application. • I spend the time doing this once. • I schedule it. • I forget about it (or at least, try to) CREAO is one of the platforms I've used extensively for building some of these automations. They are partnering with me on this post. They are one of the only platforms where you can go from a conversation to a scheduled agent that quickly. This is how it works: 1. You describe the problem you want to solve 2. CREAO's agent builds the logic and executes it 3. You can iterate with the agent to improve the solution 4. When you are done, you turn that solution into a mini app 5. You can schedule that app to run any time you need it Something important: These agents are deterministic. They don't use an LLM to generate answers, so they will always return the same output given the same input. This is critical: when you're automating something that runs every Monday at 9 am, you need it to return the same result every time, not a "creative" answer.

Right now, I have a few dozen tools to automate my life. Every single thing that I do more than once every week is now a tool: either a Claude Code skill, a scheduled workflow, or an application. • I spend the time doing this once. • I schedule it. • I forget about it (or at least, try to) CREAO is one of the platforms I've used extensively for building some of these automations. They are partnering with me on this post. They are one of the only platforms where you can go from a conversation to a scheduled agent that quickly. This is how it works: 1. You describe the problem you want to solve 2. CREAO's agent builds the logic and executes it 3. You can iterate with the agent to improve the solution 4. When you are done, you turn that solution into a mini app 5. You can schedule that app to run any time you need it Something important: These agents are deterministic. They don't use an LLM to generate answers, so they will always return the same output given the same input. This is critical: when you're automating something that runs every Monday at 9 am, you need it to return the same result every time, not a "creative" answer.

Santiago

13,712 次观看 • 4 个月前

Stop overpaying for every LLM call. 💸 The DigitalOcean Inference Router resolves intent in ~200ms and beats GPT-5.1 and Claude Sonnet 4.5 on routing accuracy. Powered by Rust, WASM, and purpose-built SLMs, it adds milliseconds, not seconds, to every request. 🌊

Stop overpaying for every LLM call. 💸 The DigitalOcean Inference Router resolves intent in ~200ms and beats GPT-5.1 and Claude Sonnet 4.5 on routing accuracy. Powered by Rust, WASM, and purpose-built SLMs, it adds milliseconds, not seconds, to every request. 🌊

DigitalOcean

19,853 次观看 • 2 个月前

You can literally tell your body what time it is by doing the exact same thing at the exact same time every day. By doing breath, work within 30 minutes of waking, we tell our circadian clock that it's morning and it's time to wake up regardless of our time zone.

You can literally tell your body what time it is by doing the exact same thing at the exact same time every day. By doing breath, work within 30 minutes of waking, we tell our circadian clock that it's morning and it's time to wake up regardless of our time zone.

Gary Brecka

78,636 次观看 • 1 年前

We've run thousands of LLM inference serving benchmarks at Modal. We're releasing the results so you don't have to. We're releasing the code so that you can. Introducing: The LLM Engineer's Almanac. Just in time for the AI Engineer World's Fair.

We've run thousands of LLM inference serving benchmarks at Modal. We're releasing the results so you don't have to. We're releasing the code so that you can. Introducing: The LLM Engineer's Almanac. Just in time for the AI Engineer World's Fair.

Charles 🎉 Frye

21,976 次观看 • 1 年前

Network related hitching / rubberband only on matchmaking servers. I no matter what i try, i can not fix it. Same console output every time the hitch happens. CS2 Fletcher Dunn

Network related hitching / rubberband only on matchmaking servers. I no matter what i try, i can not fix it. Same console output every time the hitch happens. CS2 Fletcher Dunn

elias

114,304 次观看 • 1 年前

The next iteration of Perplexity has arrived: Copilot, your interactive AI search companion. 🚀🤖 Perplexity Copilot guides your search experience with interactive inputs, leading you to a rich, personalized answer, powered by GPT-4. Try it for free at

The next iteration of Perplexity has arrived: Copilot, your interactive AI search companion. 🚀🤖 Perplexity Copilot guides your search experience with interactive inputs, leading you to a rich, personalized answer, powered by GPT-4. Try it for free at

Perplexity

1,299,853 次观看 • 3 年前

This is our second head-to-head test between Kling 3.0 and Seedance 2.0. This time, we pushed both models further, including dialogue-driven Sci-fi comedy shots and even a singing scene. Every single shot was generated using: • The exact same base image • The exact same prompt • No re-generate • No second attempt • No prompt refinements. What you see is the pure first output from each model under identical conditions. So… which one performs better? Seedance 2.0 generated by Ilham Setyawan on BytePlus

This is our second head-to-head test between Kling 3.0 and Seedance 2.0. This time, we pushed both models further, including dialogue-driven Sci-fi comedy shots and even a singing scene. Every single shot was generated using: • The exact same base image • The exact same prompt • No re-generate • No second attempt • No prompt refinements. What you see is the pure first output from each model under identical conditions. So… which one performs better? Seedance 2.0 generated by Ilham Setyawan on BytePlus

Dinda Prasetyo

196,557 次观看 • 5 个月前

When you prompt an LLM for code, you get one deterministic program. However, the LLM actually defines a distribution over many programs, and existing methods discard it‼️ PPoT uses this distribution to extract free performance and efficiency gains. 🧵👇

When you prompt an LLM for code, you get one deterministic program. However, the LLM actually defines a distribution over many programs, and existing methods discard it‼️ PPoT uses this distribution to extract free performance and efficiency gains. 🧵👇

Daniel Israel

11,655 次观看 • 3 个月前

WTF: Desperate Dems Post Videos Slamming Trump at the Exact Same Time with IDENTICAL Scripts - They're Not Beating the NPC Allegations! Tune in for more top stories here:

WTF: Desperate Dems Post Videos Slamming Trump at the Exact Same Time with IDENTICAL Scripts - They're Not Beating the NPC Allegations! Tune in for more top stories here:

Alex Jones

81,832 次观看 • 1 年前

Getting arrested by every three-letter government agency at the exact same time and still walking out after a single phone call is the ultimate boss flex in movie history.

Getting arrested by every three-letter government agency at the exact same time and still walking out after a single phone call is the ultimate boss flex in movie history.

Every Movie Plug

57,647 次观看 • 6 天前

Introducing News Analyst — a fully verifiable AI engine powered by EigenCloud suite > Uses EigenAI with wallet-signed grants. Temperature locked to zero with fixed entropy seeds. No hallucinations, only deterministic truth. > Every analysis is dispersed as a blob to EigenDA. No database—the state is stored across a validator set on Sepolia. > Generate a DA Commitment for every report. Reconstruct the entire analysis purely from the sharded network. > Sidecar architecture running via Dockerized EigenDA Proxy.

Introducing News Analyst — a fully verifiable AI engine powered by EigenCloud suite > Uses EigenAI with wallet-signed grants. Temperature locked to zero with fixed entropy seeds. No hallucinations, only deterministic truth. > Every analysis is dispersed as a blob to EigenDA. No database—the state is stored across a validator set on Sepolia. > Generate a DA Commitment for every report. Reconstruct the entire analysis purely from the sharded network. > Sidecar architecture running via Dockerized EigenDA Proxy.

Zeeshan

25,002 次观看 • 5 个月前

No ETH? No BNB? FlexGas lets you pay gas fees with TWT, USDT, or USDC. ⚙️ Same-chain now, cross-chain coming soon 🔵 Live on @BNBChain & @Ethereum 🔐 Self-custody, always. Powered by EIP-7702 Try it yourself: 🧵

No ETH? No BNB? FlexGas lets you pay gas fees with TWT, USDT, or USDC. ⚙️ Same-chain now, cross-chain coming soon 🔵 Live on @BNBChain & @Ethereum 🔐 Self-custody, always. Powered by EIP-7702 Try it yourself: 🧵

Trust Wallet

15,470 次观看 • 1 年前

Today we're excited to launch rubric scorers in Mastra The LLM-as-judge scores your agent's output as meeting (say) 7 of 11 criteria With failed checks getting fed back to the agent to re-run:

Today we're excited to launch rubric scorers in Mastra The LLM-as-judge scores your agent's output as meeting (say) 7 of 11 criteria With failed checks getting fed back to the agent to re-run:

Sam Bhagwat

13,387 次观看 • 9 天前