MathArena uses newly released competition sets and evaluates models close to the event. They also mark models released after the competition date as potential contamination.
On Feb 6, the just-concluded AIME 2026 I, Step 3.5 Flash take first place. Step 3.5 Flash was released on Feb 1, making cheating impossible.
StepFun has open-sourced Step-3.5-Flash: 196 B total parameters, 11 B active, 256 K context length.
Strong performance, with speed as the highlight—blazing fast, peaking at 350 tokens/s.
It’s currently in promotion and free on OpenRouter `step-3.5-flash:free`.
Thanks for all the great feedback! I've created a Twitter thread to discuss future development and share updates. Would love to connect with you all there:
- The DDN single-shot generator architecture is more efficient than diffusion.
- DDN is fully end-to-end differentiable, allowing for more efficient optimization when integrated with discriminative models or reinforcement learning.
Thanks for the idea, but DDN and flow can’t be flipped into each other that easily.
1. DDN doesn’t need to be invertible.
2. Its latent is discrete, not continuous.
3. As far as I know, flow keeps input and output the same size so it can compute log|detJ|. DDN’s latent is 1-D and discrete, so that condition fails.
4. To me, “hierarchical many-shot generation + split-and-prune” is simpler and more general than “invertible design + log|detJ|.”
5. Your design seems to have abandoned the characteristics of DDN. (ZSCG, 1D tree latent, lossy compression)
The two designs start from different premises and are built differently. Your proposal would change so much that whatever came out wouldn’t be DDN any more.
Even in the earliest stages of the DDN concept, we had already decided to pass features down to the next layer.
I never even ran an ablation that disabled the stem features; I assume the network would still train without them, but since the previous layer has already computed the features, it would be wasteful not to reuse them. Retaining the stem features also lets DDN adopt the more efficient single-shot-generator architecture.
Another deeper reason is that, unlike diffusion models, DDN does not need the Markov-chain property between adjacent layers.
This understanding is incorrect. The video samples all the leaf nodes of the entire tree only to visualize the distribution in latent space. In normal use, only the L outputs along a single path are generated.
That's right! The second paragraph on OPTIMIZATION WITH SPLIT-AND-PRUNE in the original paper:
> Inspired by the theory of *evolution and genetic algorithms*, we propose the Split-and-Prune algorithm to address the above issues, as outlined in algorithm 1.
I agree with @ActivePattern and thank you for your help in answering.
Supplement for @f_devd:
During training, the K outputs share the stem feature from the NN blocks, so generating the K outputs costs only a small amount of extra computation. After L2-distance sampling, discarding the other K-1 outputs therefore incurs a negligible cost and is not comparable to discarding K-1 MoE experts (which would be very expensive).
In DDN, 1×1 convolutions are used only in the output layers of the Discrete Distribution Layer (DDL).
The NN blocks between DDLs, which supply the fundamental computational power and parameter count, adopt standard 3×3 convolutions.
We provide the source code and weights along with a Docker environment to facilitate reproducing the experimental results.
The original paper’s EXPERIMENTS section mentions the hardware configuration (8× RTX 2080 Ti).
Exactly! The paragraph on Efficient Data Compression Capability in the original paper also highlights:
> To our knowledge, Taiji-DDN is the first generative model capable of directly transforming data into a semantically meaningful binary string which represents a leaf node on a balanced binary tree.
On Feb 6, the just-concluded AIME 2026 I, Step 3.5 Flash take first place. Step 3.5 Flash was released on Feb 1, making cheating impossible.