Prime Intellect Releases Multi-Agent RL Support
Users can now define and train interactions among multiple agents in prime-rl and verifiers.
Prime Intellect announced it has extended its reinforcement learning tools from single agents to multi-agent systems. Official posts state users can now program arbitrary interactions between agents, select which roles learn, and assign credit across interactions. Examples listed in the company account and engineer posts include agentic judging, self-play, user simulation, and complex agent collaboration. Research engineer samsja19 wrote that the team spent time designing the software abstraction to be extendable while staying performant. The company linked a blog post spelling out the changes in code for prime-rl and verifiers.
Today, we’re extending our RL stack beyond individual agents to multi-agent systems. You can now express arbitrary agent interactions and train them. http://primeintellect.ai/blog/multi-agent-systems
Reactions
31 postsPI 
Prime Intellect@PrimeIntellectWe believe the agent abstraction will be useful beyond training and evaluation, in particular for synthetic data pipelines. To illustrate one such agent script: set off Laguna S-2.1 to find the latest released verifiers version.
- likes: 133
- replies: 1
- bookmarks: 28
- reposts: 4
PI Prime Intellect@PrimeIntellectMulti-agent support is available today in verifiers 0.3.0 and prime-rl 0.8.0. Build your own multi-agent envs and train them with prime-rl. We’re still exploring what these abstractions make possible — and we’d love to see how far you can push them.…
- likes: 94
- replies: 0
- bookmarks: 21
- reposts: 5
PI Prime Intellect@PrimeIntellectThe core idea: an environment is just a program over agents. Run them sequentially, in parallel, or interleave them turn by turn. Mix models, harnesses, and runtimes. Multi-Agent is a natural extension of verifiers v1.
- likes: 148
- replies: 2
- bookmarks: 82
- reposts: 6
WB will brown@willcbin 2016 i got insanely nerd-sniped by multi-agent learning it was all theory at the time, but i knew that one day it would be real now it’s real
- likes: 573
- replies: 25
- bookmarks: 117
- reposts: 23
WB OK Omar Khattab@lateinteraction@willccbb what is 2016? (congrats!!)
- likes: 4
- replies: 1
- bookmarks: 0
- reposts: 0