Roger ☕☕☕

CTO / CISO Fenêtre Online Solutions - Investor - AI Fan - Husband of Judith & proud father of Mees & Ole

[Roger ☕☕☕ @rogerhendriks]

Many organizations are still searching for the best AI model in terms of quality, speed, and cost.

I believe that's the wrong question.

After benchmarking multiple AI models (@claudeai @OpenAI @Kimi_Moonshot) against real-world business software use cases, one thing became clear: there is no single "best" model. The model that excels at complex architecture or reasoning isn't always the smartest choice for bug fixes or new user stories.

The biggest gains come not from relying on a single model, but from the smart deployment of different models for different types of tasks; something we have gained considerable experience with.

That’s why we’ve shared the results of our own benchmark, including the differences in quality, speed, and cost. The models couldn't "cheat" by having been pre-trained on these specific use cases, which sometimes leads to truly surprising results.

Read the blog here: https://t.co/jPCkVHeyna

Are you already using multiple AI models side-by-side within AI agents or agentic workflows, or does your organization rely on a single standard model? Or is there a model missing from our benchmark? Feel free to let me know!

#LLMbenchmark #AIbenchmark #AI #AIdevelopment #claudebenchmark #gptbenchmark #fable

[Roger ☕☕☕ @rogerhendriks]

Another day, another model ❤️

[Google @Google]

Today we’re expanding the Gemini family with three new models built to be faster, more token efficient, and reliable at scale.

Meet the new Gemini models ↓

[Roger ☕☕☕ @rogerhendriks]

So what Saas app will you copy??

[Cursor @cursor_ai]

We had a team of agents rebuild SQLite from its 835-page manual.

It created a replica in Rust which passed 100% of a held-out test suite.

Interestingly, cost varied 15x depending on which model mix we used.

[Roger ☕☕☕ @rogerhendriks]

@AnthropicAI @OpenAI https://t.co/xxWocHqqPE

[Roger ☕☕☕ @rogerhendriks]

In our AI model test Opus 4.8 was better then Fable and Terra better than Sol. Amazing right?! See score below.
@AnthropicAI @OpenAI

The story is a kanban board that autoimproves itself with ai.

As soon as K3 will not outage anymore and we can our hands on Grok in EU we will see what happens next.

[Vals AI @ValsAI]

Kimi K3 is the #2 overall model on our in-house Vibe Code Bench at 85.0%.

VCB tests a model's ability to go from zero-to-one; creating a web application completely from scratch.

[Qwen @Alibaba_Qwen]

Qwen3.8 is launching and going open-weight soon!🌐

With a massive 2.4T parameters, this model is continuously evolving. We believe it’s one of the most powerful model available today, compatible to leading frontier AI models , second only to Fable 5.

You don't have to wait to test it. Just now, the Qwen3.8-Max-Preview made its debut on Alibaba’s Token Plan, Qoder, and QoderWork. Be among the very first to try it out.

Can't wait to hear what you build. Stay tuned! 🚀 

Token Plan
international：https://t.co/YRvcGdB9Bv
China：https://t.co/PKMUNwUuRp

[Roger ☕☕☕ @rogerhendriks]

Linus Torvalds states that AI can be superhandy so just forget all the hype, ge critical, use good tools and a professional approach 👍

[nixCraft 🐧 @nixcraft]

Linus Torvalds: “AI is tool and will be used in Linux kernel development, anyone who’s is not happy about it can fork Linux kernel” 🤣

[Roger ☕☕☕ @rogerhendriks]

@iruletheworldmo We will put it against our business software benchmark with no option for pretraining. Let's see if it's fake or real.. https://t.co/PW371Fs1vI

[Roger ☕☕☕ @rogerhendriks]

Many organizations are still searching for the best AI model in terms of quality, speed, and cost.

I believe that's the wrong question.

After benchmarking multiple AI models (@claudeai @OpenAI @Kimi_Moonshot) against real-world business software use cases, one thing became clear: there is no single "best" model. The model that excels at complex architecture or reasoning isn't always the smartest choice for bug fixes or new user stories.

The biggest gains come not from relying on a single model, but from the smart deployment of different models for different types of tasks; something we have gained considerable experience with.

That’s why we’ve shared the results of our own benchmark, including the differences in quality, speed, and cost. The models couldn't "cheat" by having been pre-trained on these specific use cases, which sometimes leads to truly surprising results.

Read the blog here: https://t.co/jPCkVHeyna

Are you already using multiple AI models side-by-side within AI agents or agentic workflows, or does your organization rely on a single standard model? Or is there a model missing from our benchmark? Feel free to let me know!

#LLMbenchmark #AIbenchmark #AI #AIdevelopment #claudebenchmark #gptbenchmark #fable

[Roger ☕☕☕ @rogerhendriks]

Can't wait to put it in our benchmark 👍
https://t.co/PW371Fs1vI

[leo 🐾 @synthwavedd]

🚨 BREAKING: Kimi K3 benchmarks have released and it's ~Fable/Sol level

[Roger ☕☕☕ @rogerhendriks]

@aidenybai Interesting to see the differences/similarities! we ran our own benchmark as well, you can check it out here https://t.co/IqUWO5iqkF

[Roger ☕☕☕ @rogerhendriks]

Many organizations are still searching for the best AI model in terms of quality, speed, and cost.

I believe that's the wrong question.

After benchmarking multiple AI models (@claudeai @OpenAI @Kimi_Moonshot) against real-world business software use cases, one thing became clear: there is no single "best" model. The model that excels at complex architecture or reasoning isn't always the smartest choice for bug fixes or new user stories.

The biggest gains come not from relying on a single model, but from the smart deployment of different models for different types of tasks; something we have gained considerable experience with.

That’s why we’ve shared the results of our own benchmark, including the differences in quality, speed, and cost. The models couldn't "cheat" by having been pre-trained on these specific use cases, which sometimes leads to truly surprising results.

Read the blog here: https://t.co/jPCkVHeyna

Are you already using multiple AI models side-by-side within AI agents or agentic workflows, or does your organization rely on a single standard model? Or is there a model missing from our benchmark? Feel free to let me know!

#LLMbenchmark #AIbenchmark #AI #AIdevelopment #claudebenchmark #gptbenchmark #fable

[Roger ☕☕☕ @rogerhendriks]

@Kimi_Moonshot For who's interested: we benchmarked several models with a real use-case, including Kimi K2.7. Check it out: https://t.co/IqUWO5iqkF

[Roger ☕☕☕ @rogerhendriks]

Many organizations are still searching for the best AI model in terms of quality, speed, and cost.

I believe that's the wrong question.

After benchmarking multiple AI models (@claudeai @OpenAI @Kimi_Moonshot) against real-world business software use cases, one thing became clear: there is no single "best" model. The model that excels at complex architecture or reasoning isn't always the smartest choice for bug fixes or new user stories.

The biggest gains come not from relying on a single model, but from the smart deployment of different models for different types of tasks; something we have gained considerable experience with.

That’s why we’ve shared the results of our own benchmark, including the differences in quality, speed, and cost. The models couldn't "cheat" by having been pre-trained on these specific use cases, which sometimes leads to truly surprising results.

Read the blog here: https://t.co/jPCkVHeyna

Are you already using multiple AI models side-by-side within AI agents or agentic workflows, or does your organization rely on a single standard model? Or is there a model missing from our benchmark? Feel free to let me know!

#LLMbenchmark #AIbenchmark #AI #AIdevelopment #claudebenchmark #gptbenchmark #fable

[Roger ☕☕☕ @rogerhendriks]

@bridgemindai @bridgebench If you're interested in AI benchmarks: We benchmarked several models with a real use-case. Read more here: https://t.co/Re0ISZlYlY

[Roger ☕☕☕ @rogerhendriks]

Many organizations are still searching for the best AI model in terms of quality, speed, and cost.

I believe that's the wrong question.

After benchmarking multiple AI models (@claudeai @OpenAI @Kimi_Moonshot) against real-world business software use cases, one thing became clear: there is no single "best" model. The model that excels at complex architecture or reasoning isn't always the smartest choice for bug fixes or new user stories.

The biggest gains come not from relying on a single model, but from the smart deployment of different models for different types of tasks; something we have gained considerable experience with.

That’s why we’ve shared the results of our own benchmark, including the differences in quality, speed, and cost. The models couldn't "cheat" by having been pre-trained on these specific use cases, which sometimes leads to truly surprising results.

Read the blog here: https://t.co/jPCkVHeyna

Are you already using multiple AI models side-by-side within AI agents or agentic workflows, or does your organization rely on a single standard model? Or is there a model missing from our benchmark? Feel free to let me know!

#LLMbenchmark #AIbenchmark #AI #AIdevelopment #claudebenchmark #gptbenchmark #fable

[Roger ☕☕☕ @rogerhendriks]

@claudeai We benchmarked several models with a real use-case, including this one! Check it out here: https://t.co/Re0ISZlYlY

[Roger ☕☕☕ @rogerhendriks]

Many organizations are still searching for the best AI model in terms of quality, speed, and cost.

I believe that's the wrong question.

After benchmarking multiple AI models (@claudeai @OpenAI @Kimi_Moonshot) against real-world business software use cases, one thing became clear: there is no single "best" model. The model that excels at complex architecture or reasoning isn't always the smartest choice for bug fixes or new user stories.

The biggest gains come not from relying on a single model, but from the smart deployment of different models for different types of tasks; something we have gained considerable experience with.

That’s why we’ve shared the results of our own benchmark, including the differences in quality, speed, and cost. The models couldn't "cheat" by having been pre-trained on these specific use cases, which sometimes leads to truly surprising results.

Read the blog here: https://t.co/jPCkVHeyna

Are you already using multiple AI models side-by-side within AI agents or agentic workflows, or does your organization rely on a single standard model? Or is there a model missing from our benchmark? Feel free to let me know!

#LLMbenchmark #AIbenchmark #AI #AIdevelopment #claudebenchmark #gptbenchmark #fable

[Roger ☕☕☕ @rogerhendriks]

@OpenAI We benchmarked several models including this one! check it out here: https://t.co/Re0ISZlYlY

[Roger ☕☕☕ @rogerhendriks]

Many organizations are still searching for the best AI model in terms of quality, speed, and cost.

I believe that's the wrong question.

After benchmarking multiple AI models (@claudeai @OpenAI @Kimi_Moonshot) against real-world business software use cases, one thing became clear: there is no single "best" model. The model that excels at complex architecture or reasoning isn't always the smartest choice for bug fixes or new user stories.

The biggest gains come not from relying on a single model, but from the smart deployment of different models for different types of tasks; something we have gained considerable experience with.

That’s why we’ve shared the results of our own benchmark, including the differences in quality, speed, and cost. The models couldn't "cheat" by having been pre-trained on these specific use cases, which sometimes leads to truly surprising results.

Read the blog here: https://t.co/jPCkVHeyna

Are you already using multiple AI models side-by-side within AI agents or agentic workflows, or does your organization rely on a single standard model? Or is there a model missing from our benchmark? Feel free to let me know!

#LLMbenchmark #AIbenchmark #AI #AIdevelopment #claudebenchmark #gptbenchmark #fable

[NXT EU @NXT4EU]

Germany has launched one of the world's best open-source AI models.

Soofi S, made by the Soofi consortium, is a 30B parameter model fully trained in Europe and tops the ranking for open-source AI.

Huge moment for Europe, and finally some competition for Chinese open-source AI.

[Vals AI @ValsAI]

The much-anticipated Grok 4.5 is here, landing at #6 on the Vals Index (65.3%) and nearly 20 percentage points better than Grok 4.3.

[Fenêtre Online Solutions @FenetreBV]

For #ScrumMasters and #DevOps teams using #AzureDevOps:

Sprint templates, task linking, tag cleanup and PR monitoring: It can be easier than you think!

We built DevOps InControl for people like you, a devops management tool, free & open source.

Try it https://t.co/iZVKbwSW4k

[🍓🍓🍓 @iruletheworldmo]

‼️breaking

>fable 5 will be fully restored this week
>breakthrough in talks today directly with dario and us gov
>expecting news to officially break in the next 48hrs.