Tyler Griggs

@tyler_griggs_

@thinkymachines, PhD @UCBerkeley Sky Lab building SkyRL

Berkeley, CA

tyler-griggs.github.io

Replies Reposts Media

Tyler Griggs
@tyler_griggs_
Jul 15
Very happy to share Inkling with the world
Thinking Machines
@thinkymachines
Jul 15
Today, we are introducing Inkling. Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available. thinkingmachines.ai/news/introduci… Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵
6
Tyler Griggs
@tyler_griggs_
May 12
For all the progress in model intelligence, we're still in the early days of interactivity. There are definitely better modes of interaction than turn-based exchanges that freeze reality while the model generates. See the team's research preview that explores what's next:
Thinking Machines
@thinkymachines
May 11
People talk, listen, watch, think, and collaborate at the same time, in real time. We've designed an AI that works with people the same way. We share our approach, early results, and a quick look at our model in action. thinkingmachines.ai/blog/interacti…
00:00
Tyler Griggs
@tyler_griggs_
Feb 19
🥹
Hao Kang
@Mlsys_HaoKang
Feb 19
Replying to @NovaSkyAI
Personally, SkyRL is the best code repo for agentic RL. We have tested lots of other RL codebase and their examples can not even run correctly. Respect!
Tyler Griggs
@tyler_griggs_
Feb 17
Harbor has increasingly become a standard for agentic evals. Now you can train in the same format! Great work by @charlie_ruan and the team to get SkyRL and Harbor running together smoothly
Charlie Ruan
@charlie_ruan
Feb 17
Releasing the official SkyRL + Harbor integration: a standardized way to train terminal-use agents with RL. From the creators of Terminal-Bench, Harbor is a widely adopted framework for evaluating terminal-use agents on any task expressible as a Dockerfile + instruction + test
1
Tyler Griggs
@tyler_griggs_
Feb 13
SkyRL now implements the Tinker API. Now, training scripts written for Tinker can run on your own GPUs with zero code changes using SkyRL's FSDP2, Megatron, and vLLM backends. Blog: novasky-ai.notion.site/skyrl-tinker 🧵
6