Agentic Intelligence Lab retweeted
huggingface.co/spaces/vllm-s…
Checkout the d3 turbo HF space 🤗 collected some cool demos 👻
Agentic Intelligence Lab retweeted
Open d3-turbo crushed Jev 1.14 at Whack-a-Duck 🦆
@vllm_project d3-turbo on a single RTX PRO 6000 decides in 258 ms so its mallet hit 32 of 34 ducks while @typesafeai Jev 1.14 took 414 ms a call over the API and hit 21
Run decision models locally -> atomic.chat
Introducing Decision 3.0: towards open multimodal foundation decision models ⚡
Open and state-of-the-art from 0.6B to 27B: #1 on both Jev Decision Index boards in our evaluation, on a new Pareto frontier.
HF collection 🤗: huggingface.co/collections/v…
#OpenSource #Jev #LLM
Agentic Intelligence Lab retweeted
Introducing d3-turbo: our newest state-of-the-art multimodal foundation decision model⚡
#1 on both Jev Decision Index boards. It points(box), follows and sees images in your questions. StateOnce brings 15x faster than d3 flash
🤗 huggingface.co/vllm-sr/d3-tu…
#vLLM #Jev #Decision
Agentic Intelligence Lab retweeted
Turn any camera into a stream of decisions.
Every second, d3-edge (0.6B) watched this rescue and answered 8 plain-English questions in ~0.1 s.
New questions, new scene: pet cam, bird feeder, doorstep. What would you ask?
huggingface.co/vllm-sr/d3-ed…
#Jev #Decision
Agentic Intelligence Lab retweeted
Introduce Decision 3.0 Free API 🚀 for developers to learn, research and explore what decision models can do
#1 on both Jev Decision Index. Six models live, 0.6B to 27B. Text, images or video in, a probability for every answer out
One click, no account. SystemOne API compatible
Agentic Intelligence Lab retweeted
d3-edge on a Mac, from a parkour head cam 🏃
Every 2 s of footage, 8 plain yes/no questions up high? long drop below? holding on? on steel? answered in ~110 ms on MLX, labels drawn live on the video with some music.
Core ML + MLX ports 🤗 huggingface.co/FluidInferenc…
Turn any camera into a stream of decisions.
Every second, d3-edge (0.6B) watched this rescue and answered 8 plain-English questions in ~0.1 s.
New questions, new scene: pet cam, bird feeder, doorstep. What would you ask?
huggingface.co/vllm-sr/d3-ed…
#Jev #Decision
Agentic Intelligence Lab retweeted
Decision 3.0 hype is real! it is the top performer in my little apple-harvesting test. It can actually play: read the scene, pick its own moves and collect apples without a pathfinding controller.
Input: a screenshot of its surroundings, game stats and a short memory of its last four moves. It chooses up/down/left/right each turn.
This run: 7 apples in 38 moves, zero damage 🔥
Introducing Decision 3.0: towards open multimodal foundation decision models ⚡
Open and state-of-the-art from 0.6B to 27B: #1 on both Jev Decision Index boards in our evaluation, on a new Pareto frontier.
HF collection 🤗: huggingface.co/collections/v…
#OpenSource #Jev #LLM
Agentic Intelligence Lab retweeted
A few days ago @XunzhuoLiu reached out with early access to D3, I didn't expect to end it with:
- a blog post vllm-sr.ai/blog/decision-3-0… 🥳
- joining 2 Semantic Router workgroups (Router Models & Inference Runtime, DX & Ecosystem) vllm-sr.ai/community/work-gr… 🎉
- @ClementDelangue to repost my d3 coreml post 🤯
Thank you everyone for the trust! To celebrate here's d3, distilled and on Core ML, playing Super Mario on a M5 Pro 🍄
Agentic Intelligence Lab retweeted
let's go huggingface.co/vllm-sr/d3
Agentic Intelligence Lab retweeted
All sizes of d3 models published at official jev decision index, and 5 of 6 released models (except the smallest 0.6B , also top2 at the same size), reached at the pareto frontier.
This means, no matter which model you wanto to pick, it is always the best at the same size. 🪄
Agentic Intelligence Lab retweeted
Our d3 is officially top #1 at the Jev Decision Index 0.3.1 now 🔥 This is the moment for open source and the ecosystem. It’s the reason why I love building in open, since great minds meet together, more advancing things will happen. Let’s go for the new SOTA 🤗
#Jev #Decision
Agentic Intelligence Lab retweeted
Officially d3 is the new top now! This is the celebration to the open source ecosystem, power of the community and great minds meet together to push the new SOTA 🤗
xitter.cf/multimodalart/status/2…
New models live on Decision Index 0.3.1 🏆
vLLM Decision 3.0 (d3) & Mapika Decider 31B tie #1 🏅
d3 uses Perplexity's Decider arch (now tied #2 🥈)
AnyJev Gemma-4-31B (tied #3 🥉) is the first inference technique (untuned model) above Jev, tied #4 🎖️
huggingface.co/spaces/multim…
Agentic Intelligence Lab retweeted
New models live on Decision Index 0.3.1 🏆
vLLM Decision 3.0 (d3) & Mapika Decider 31B tie #1 🏅
d3 uses Perplexity's Decider arch (now tied #2 🥈)
AnyJev Gemma-4-31B (tied #3 🥉) is the first inference technique (untuned model) above Jev, tied #4 🎖️
huggingface.co/spaces/multim…
Agentic Intelligence Lab retweeted
Decision 3.0 now on Core ML 🍎
d3-lite triaging 1,000 pull requests into vllm-sr workgroups, live on an M5 Pro: 0.12 s per PR, 8.3 PRs/s, nothing leaves the Mac ⚡
lite, nano and mini are all converted and in one repo 👇
🤗 huggingface.co/FluidInferenc…
🛠️ github.com/FluidInference/Fl…
Introducing Decision 3.0: towards open multimodal foundation decision models ⚡
Open and state-of-the-art from 0.6B to 27B: #1 on both Jev Decision Index boards in our evaluation, on a new Pareto frontier.
HF collection 🤗: huggingface.co/collections/v…
#OpenSource #Jev #LLM
Agentic Intelligence Lab retweeted
It is a vast improvement over d2! coreml repo + fluiduse sdk + speed coreml demo coming later!
Introducing Decision 3.0: towards open multimodal foundation decision models ⚡
Open and state-of-the-art from 0.6B to 27B: #1 on both Jev Decision Index boards in our evaluation, on a new Pareto frontier.
HF collection 🤗: huggingface.co/collections/v…
#OpenSource #Jev #LLM
Agentic Intelligence Lab retweeted
Introducing Decision 3.0: towards open multimodal foundation decision models ⚡
Open and state-of-the-art from 0.6B to 27B: #1 on both Jev Decision Index boards in our evaluation, on a new Pareto frontier.
HF collection 🤗: huggingface.co/collections/v…
#OpenSource #Jev #LLM