@samsja19: prime-rl can now train 1T parameters MoE blazingly fast, under 5 minutes per step, or 1k steps in ~3 days To achieve th…
Summary
Prime Intellect released prime-rl v0.6.0, enabling reinforcement learning at trillion-parameter MoE scale with sub-5-minute step times and optimized inference, training, and rollout.
View Cached Full Text
Cached at: 06/23/26, 03:52 PM
prime-rl can now train 1T parameters MoE blazingly fast, under 5 minutes per step, or 1k steps in ~3 days
To achieve this we shipped in our latest prime-rl 0.6.0:
-
inference: wide-ep, fp8 inference, llm-d router, mooncake, kv cache cpu offloading
-
training: fsdp2, deep-ep expert parallelism, dsa cp, fp8 training, router replay
-
agentic rollout: we rewrote the core of our rollout orchestrator for better scalability
support for glm5, kimi, nemotron, …,
prime-rl is open source but also end to end optimized to run on our dedicated RL infra and compute layer
Prime Intellect (@PrimeIntellect): Today we’re releasing prime-rl v0.6.0 — enabling RL at trillion-parameter MoE scale on agentic workloads at the highest efficiency.
We’ve relentlessly optimized our RL infra.
The result: GLM-5 on agentic SWE tasks at 131k context and sub-5-minute step time.
Similar Articles
@eliebakouch: every infra piece you need to know to do RL on GLM-5 https://primeintellect.ai/blog/rl-at-1t-scale…
Prime Intellect releases prime-rl v0.6.0, enabling efficient reinforcement learning at trillion-parameter scale on large Mixture-of-Experts models, with sub-5-minute step times and optimizations for asynchronous RL.
@samsja19: with multi agent also come prime-rl 0.8.0 release, 98 commits from 13 contributors since 0.7.0 on the menu: 1. multi ag…
Prime Intellect releases prime-rl 0.8.0, adding multi-agent training, Nixl and model express weight broadcast, extended multimodal support, and various performance improvements.
@h100envy: Prime Intellect engineers explained how they train reasoning models over the open internet in 30 minutes - better than …
Prime Intellect engineers demonstrated a method to train reasoning models in 30 minutes using distributed RL over the open internet, utilizing Prime-RL, LLM judges, and multi-cloud GPUs, enabling open models to compete with closed labs without owning data centers.
@samsja19: https://x.com/samsja19/status/2076846033922035818
PRIME-RL is a framework for large-scale asynchronous reinforcement learning, designed to be hackable and scale to 1000+ GPUs with support for various models and environments.
@samsja19: We are also releasing prime-rl 0.7.0 which has full support for verifiers v1 and bring your own harness for training. W…
Prime Intellect released verifiers v1 and prime-rl 0.7.0, an RL training tool with full support for verifiers, multiple algorithms like GRPO and OPD, and performance improvements.