Open Superintelligence Factory.
Joined June 2020
- Tweets3.2K
- Following49
- Followers88K
- Likes44.4K
Open Superintelligence Research Day
Friday, October 23
Berlin
Own your intelligence
Own the factory that improves it
Own the means of production
All of this runs on one eval stack. Run evals in prime-rl: any model, any environment, either standalone or as online evals during a run
verifiers 0.4.0 ships alongside. Everything is v1 (v0 removed) and the command-line tools are all vf-* now
Every architecture in prime-rl now runs on our own optimized implementation
New in this release: DeepSeek V4 Flash, Qwen3.8 Flash Next, Nemotron-H Nano and Super, and multimodal RL with Qwen-VL and Kimi K2.5
We tested it in a deliberately hard setting: agentic SWE RL with GLM-4.5-Air
• IcePop collapsed around step 1.2k when trainer-inference KL blew up 20×
• IPO trained straight through, masking only ~0.25% of tokens
INTELLECT Policy Optimization (IPO)
In async RL, trainer and inference never agree exactly on a token's probability. IcePop masks on the ratio, which misjudges both tails
IPO masks on the probability difference instead
With GLM-5.3, sparse attention now runs the forward pass with FlashMLA and the backward with cuDNN, plus a 12-node 16k recipe with no context parallelism or offloading
On full SFT, throughput is 1.8× our previous recipe
Faster training
Qwen3-235B-A22B SFT on the same nodes as torchtitan (4×8 B300, 16k seq) is 1.15× faster with the same memory
Most of the gain is MoE work: fused gate/up and QKV projections, fullgraph compile, chunked loss, and SonicMoE BF16 backend
Today we're releasing prime-rl v0.10.0 and verifiers v0.4.0
We added our new RL algorithm IPO as the default, tailored for async RL stability, as well as performance improvements and new model support
Sword Health built Dawn 1 on prime-rl:
They RL-post-trained GLM-5.3 into a mental health support model that is safer, less sycophantic, and more cost-efficient than its base
Introducing Dawn 1, Sword's frontier model for mental health support.
Post-trained from GLM-5.3, Dawn 1 delivers greater intelligence at lower inference cost, with strong performance across mental health knowledge, safety, and sycophancy evaluations. It also outperforms frontier proprietary models on multi-turn safety protocol adherence in our evaluations.
A look at the research 🧵
We validated tasks on verifiers, renderers, and prime-rl
NVIDIA used these verified tasks to train Nemotron 3.5 Super VL with GRPO
This offers a path to keep improving enterprise agents: use workflow failures to build the next set of verified tasks, then post-train and evaluate again
Together with NVIDIA, we built 2,500 hard enterprise tasks modeled on AutomationBench
We tested for shortcuts and answer leakage and ensured that graders reward valid solutions without accepting incomplete work
Enterprise agents need to find context across applications, reason through policies, and complete work over multiple turns
AutomationBench evaluates agents across domains in simulated SaaS applications ranging from CRMs to inboxes
Here's how we increased the score from 34.0% to 57.7%
