Lorenzo Porras

Founding AI Engineer, Developer Relations @Wandb @Coreweave 🐝✨
Content on Serverless Inference, Evals, Sandbox the stuff that’s actually moving right now

[Lorenzo Porras @TheCodingSoup]

During @reflection_ai’s private preview of Beam, I built a haunted escape room to put its coding and reasoning to the test.

I created a game called “Code Your Way Out”, and it's one room with three linked puzzles. Beam has to repair a brass orrery by lining its rings up with a constellation, route the light, then program its escape through a sealed hatch.

Beam never sees the room. Each turn, it gets text observations and writes Python; the code runs in a @CoreWeave Sandbox, and the game engine validates every action before the explorer robot moves.

Beam was out in about 30 seconds. Super bullish on this model and really excited to see builders get this model in their hands.

[CoreWeave @CoreWeave]

🔦 Beam, Reflection's first model with open weights, is live today on CoreWeave Serverless Inference.

Call it through an API and leave the GPUs to us.

We are also proud to share that @reflection_ai used @wandb in CoreWeave Forge during Beam's training.

Developers in the private preview have been calling it fast and accurate, and they like that it doesn't ramble.

See what @TheCodingSoup on our team built with private preview access below.

Get started with Beam here: https://t.co/tyRVazUtUm

[Lorenzo Porras @TheCodingSoup]

@CoreWeave @reflection_ai @wandb https://t.co/1NN88eRj2t

[Lorenzo Porras @TheCodingSoup]

During @reflection_ai’s private preview of Beam, I built a haunted escape room to put its coding and reasoning to the test.

I created a game called “Code Your Way Out”, and it's one room with three linked puzzles. Beam has to repair a brass orrery by lining its rings up with a constellation, route the light, then program its escape through a sealed hatch.

Beam never sees the room. Each turn, it gets text observations and writes Python; the code runs in a @CoreWeave Sandbox, and the game engine validates every action before the explorer robot moves.

Beam was out in about 30 seconds. Super bullish on this model and really excited to see builders get this model in their hands.

[Lorenzo Porras @TheCodingSoup]

Its out!

Try @reflection_ai Beam on @CoreWeave serverless inference!

If you also get a chance check out the demo I built that is on the launch page, and pinned on my X as well!

[CoreWeave @CoreWeave]

🔦 Beam, Reflection's first model with open weights, is live today on CoreWeave Serverless Inference.

Call it through an API and leave the GPUs to us.

We are also proud to share that @reflection_ai used @wandb in CoreWeave Forge during Beam's training.

Developers in the private preview have been calling it fast and accurate, and they like that it doesn't ramble.

See what @TheCodingSoup on our team built with private preview access below.

Get started with Beam here: https://t.co/tyRVazUtUm

[Lorenzo Porras @TheCodingSoup]

��👀👀

[Weights & Biases @wandb]

New model dropping soon on @CoreWeave Serverless Inference 👀

CoreWeave Agent Lens (@weave_wb) tracing built in.

[Lorenzo Porras @TheCodingSoup]

Interesting rundown of where benchmarking stands and what benchmaxxing actually looks like in practice.

[Quanta Magazine @QuantaMagazine]

Moritz Hardt thinks that OpenAI and Anthropic’s claims about benchmarks, which are used to test AI models’ capabilities, are misleading. But if the tests aren’t accurate, then how do we know how good the models really are? Read the interview: https://t.co/0nCHaMIKUW

[Lorenzo Porras @TheCodingSoup]

@peer_rich I mean some things just need to remain untouched.

[Lorenzo Porras @TheCodingSoup]

@reflection_ai just launched their new model, Beam, and my demo made it onto their launch blog! 🎉

For the demo, I built Beam an escape room. I locked it inside a series of puzzles and let it work through each problem on its own to find a way out.

Watching it reason through the clues, test ideas, and adjust when something didn’t work was a fun way to see what the model can do. Tracked everything using @CoreWeave sandbox and agent lens!

Congrats to the Reflection team on the launch. Check out the blog to see the demo in action.

https://t.co/qp3vRqmsJz

[Lorenzo Porras @TheCodingSoup]

Beam is a lot of fun to work with. Checkout my pinned post on how I used it for an escape room!

[Reflection @reflection_ai]

Beam is now available across platforms, with open weights under the Apache 2.0 license - in collaboration with an ecosystem of 40+ partners.

We're releasing:
- Weights, tokenizer, and reference implementation
- BF16, FP8 and NVFP4 weights and a DFlash2 drafter
- The open source harness we built and trained Beam on: Mirror CLI

Weights and model card on Hugging Face: https://t.co/tPqWpk9RXL

[Lorenzo Porras @TheCodingSoup]

https://t.co/M3yaRWZfZS

[Reflection @reflection_ai]

Beam is now available across platforms, with open weights under the Apache 2.0 license - in collaboration with an ecosystem of 40+ partners.

We're releasing:
- Weights, tokenizer, and reference implementation
- BF16, FP8 and NVFP4 weights and a DFlash2 drafter
- The open source harness we built and trained Beam on: Mirror CLI

Weights and model card on Hugging Face: https://t.co/tPqWpk9RXL

[Lorenzo Porras @TheCodingSoup]

@reflection_ai https://t.co/M3yaRWZfZS

[Lorenzo Porras @TheCodingSoup]

During @reflection_ai’s private preview of Beam, I built a haunted escape room to put its coding and reasoning to the test.

I created a game called “Code Your Way Out”, and it's one room with three linked puzzles. Beam has to repair a brass orrery by lining its rings up with a constellation, route the light, then program its escape through a sealed hatch.

Beam never sees the room. Each turn, it gets text observations and writes Python; the code runs in a @CoreWeave Sandbox, and the game engine validates every action before the explorer robot moves.

Beam was out in about 30 seconds. Super bullish on this model and really excited to see builders get this model in their hands.

[Reflection @reflection_ai]

Beam is now available across platforms, with open weights under the Apache 2.0 license - in collaboration with an ecosystem of 40+ partners.

We're releasing:
- Weights, tokenizer, and reference implementation
- BF16, FP8 and NVFP4 weights and a DFlash2 drafter
- The open source harness we built and trained Beam on: Mirror CLI

Weights and model card on Hugging Face: https://t.co/tPqWpk9RXL

[Chetan Tekur @chetantekur]

Beam weights have dropped along with incredible ecosystem support! What are you going to build with Beam! Lessgo! 🚀🚀🚀🚀

[AI at Meta @AIatMeta]

Today we’re releasing Muse Spark Open, the most capable US open-weight model to date.

Built for coding and agentic work, it excels at long-horizon tasks, complex software engineering, and multi-step reasoning.

Key Model Specs:
→ 816B total / 42B active parameters
→ 1M token context window
→ Natively multimodal across text, image, video, audio

[Lorenzo Porras @TheCodingSoup]

Finally... 😭😭😭

I'm so sick of juggling sessions across 10 terminal tabs with zero idea what's running where side by side might actually make the desktop app usable.

[ClaudeDevs @ClaudeDevs]

We've been working hard on making the Claude Code desktop app great for running lots of sessions side by side.

It's now much quicker to start sessions, easier to split work across a few, and you can always see what each of them is doing. Here’s what's new:

[Lorenzo Porras @TheCodingSoup]

And yet another model releases 👀

[Daanish Khazi @bertgodel]

We are releasing Ultramarine 1 Flash, a frontier-class model built for knowledge work. It is open-source and achieves similar performance to Fable 5, while being 60x cheaper.

It was post-trained by a team of 5 for ~$300k all-in. Available on API now, with weights releasing in 2 weeks.

[Lorenzo Porras @TheCodingSoup]

@CoreWeave @deok_filho Check out the inspiration from @ktaletsk
https://t.co/5y8vtNAvXp

[Lorenzo Porras @TheCodingSoup]

Gave an AI five functions and a tiny island.

It walked into a hole.

Now it has to collect the gems, build the bridge, and reach the portal.

Running on @CoreWeave Sandbox: one deploy script, a public endpoint, set to stay up for 30 days.

https://t.co/m8thdUDA7k

Which model would you try? 👇

[Deok @deok_filho]

Just handoff all your agent sessions to a cloud sandbox with 'cws-agent handoff'. This has saved me many times on long sessions that I wanted to continue working on as I left the house.
Close that laptop!

[Lorenzo Porras @TheCodingSoup]

@altryne @Meta @Muse Done ✅

[Lorenzo Porras @TheCodingSoup]

Trying out @Meta @Muse for the first time. I’ve not had an AI personal assistant before and heard great things.

Really enjoying it so far, setting up goals, prepping content, fitness goals as well.

I need to find more ways to make this useful!

Any suggestions??