Jerry Hong

post training @ baseten

[Jerry Hong @jerryhong21]

https://t.co/T9qBi0R7ph

[Jerry Hong @jerryhong21]

@oneill_c incredible

[Jerry Hong @jerryhong21]

So so exciting to see this launch. The team's delivered something phenomenal.

To epitomize how relentless the team has worked, last week our intern ask an hour off to go to a concert, on a saturday.

[Baseten @baseten]

Baseten Loops is now GA, and serverless.

Teams at @AmbienceAI, @datadoghq, and @LangChain use the Loops SDK for SFT and RL training to build durable intelligence.

Serverless Loops includes:
- Always-warm infra that eliminates cold starts
- Pay-per-token pricing for training and sampling
- One-click deployment from your trained checkpoint to production inference
- Model-specific numerical validation, with a focus on reducing mismatch between training and sampling

You bring the data, reward functions, and training logic.
We handle the infrastructure.
You always own the weights.

Learn more: https://t.co/vA6PvHEJq2

[Jerry Hong @jerryhong21]

this is what happens when obsessed, independent researchers reach flow state. awesome results from @jacknzheng and @jeffreycider

[cider @jeffreycider]

we discovered the edge of stability for muon
(and also every other non-adaptive optimizer)

we call it bubbleforce.

also we shaved off 100 iterations from the nanoGPT muon record (-3.2%)

[Jerry Hong @jerryhong21]

@IKorovinsky ian k3.1 flash dropping soon 🔥🔥🔥

[Jerry Hong @jerryhong21]

https://t.co/T9qBi0R7ph

[Jerry Hong @jerryhong21]

@waterloo_intern as an adjacent rule:

- if you need something optimized, go talk to to @waterloo_intern

[Jerry Hong @jerryhong21]

@IKorovinsky @mudithj @stefanopopoulos he doesn’t even speak english😭😭

[Jerry Hong @jerryhong21]

@oneill_c @stefanopopoulos i think you misspelt BLT

[Jerry Hong @jerryhong21]

5 months ago, I didn’t know what an MoE layer was, now, I kinda do.

Anyways, here’s how @stefanopopoulos and I made the fastest delta weight syncs in the world.

[Jerry Hong @jerryhong21]

https://t.co/T9qBi0R7ph

[Jerry Hong @jerryhong21]

@IKorovinsky @stefanopopoulos i’m pretty sure the router will direct you to the top-k chocolate layers

[Jerry Hong @jerryhong21]

BRRRRRRRRR

[Base Labs @baselabs]

Today we release mammoth-1 and elephant-1!

We post-trained mammoth-1, the #1 ranked best decision model in the world, and elephant-1, which is the best model under 15B (and beats Jev).

We're open-sourcing all model weights and provide a technical report detailing all our training and data methods.

[Base Labs @baselabs]

Today we release mammoth-1 and elephant-1!

We post-trained mammoth-1, the #1 ranked best decision model in the world, and elephant-1, which is the best model under 15B (and beats Jev).

We're open-sourcing all model weights and provide a technical report detailing all our training and data methods.

[Chris Yoo @pentestduck]

Today we're incredibly excited to announce @DreamscaleLabs (YC F26)! We're an applied research company putting robot brains in the cloud.

Robot AI models are growing larger and on-board compute is getting increasingly inaccessible due to cost. We're taking a contrarian bet that the future is moving inference off the robot and into the cloud, while still meeting real-time deadlines required by robot motion.

We're building the real-time inference infrastructure that enables frontier physical AI models to run smoothly on your robot. We optimize for low latency across the full stack: from the model, down to the GPU kernels, and over the network.

Our mission is to unlock a future of safe and intelligent robots doing meaningful work, and we think this is only possible by letting you deploy and run robot models in the cloud.

Reach out to us at [email protected] or visit our website at https://t.co/AFAJApNM2H to learn more.

@KDog112358 @AntoineNeedGPUs @eternaI_entropy

[Jerry Hong @jerryhong21]

@IKorovinsky @GarageCapital @inovia @TechyonOrg @HackTheNorth @wlootechweek freshhh

[Jerry Hong @jerryhong21]

rumour is that @ilyasut came to @jacknzheng for advice at his lowest point

[Jack Neo Zheng @jacknzheng]

We just broke 2 NanoGPT world records

On track 3 optimization, we set 2 new world records on both the Muon and MuonH optimizers by 90 (-3%) and 60 (-2%) steps.

We implement a simple variant of single EMA momentum - using multiple EMA buffers and anneal momentum during training

[Jerry Hong @jerryhong21]

@IKorovinsky also semi pro fortnite player

[Jerry Hong @jerryhong21]

did anyone mention he can also do 12 reps of 45lb weighted pull ups 💚

[Paras Stefanopoulos @stefanopopoulos]

Sub 40s delta weight syncs for GLM 5.3 🔥💚 https://t.co/0rBdvgMjCw

[Jerry Hong @jerryhong21]

@suganton lost the plot 🥀🥀🥀

[Jerry Hong @jerryhong21]

brace yourselves 🔥

[Base Labs @baselabs]

Today we're announcing Base Labs, a dedicated research organization focused on advancing open-source AI.

We believe in a healthy, open frontier model ecosystem. To enable this, we are working on:

- Blue-sky research on continual learning, the science of RL, and how models learn across their full lifecycle, with every experiment and recipe shared openly.

- The BaseHub Data Foundry: the highest-quality open RL environments, training data, and real-world benchmarks, built for anyone to train and benchmark on.

- Post-post training: taking open-source models and making them better, safer, and more aligned through continual post-training, built on our research, and deployed with our frontier safety stack so organizations can use open-source models with confidence.

- Making models cheaper and more performant through our model performance research.

This is a mission-driven research effort, not a commercial product. We believe the health of the open-source AI ecosystem matters and that the best way to advance it is to do science in the open.

We’re hiring engineers, researchers, and research fellows to advance this mission.

https://t.co/Go8YsfaP2X

[Jerry Hong @jerryhong21]

specimen 🔥

[Brian Li @BrianLi23]

https://t.co/ZyhZUElPGk