Prime Intellect

2026•Prime Intellect

Lab v1.0

Lab is a comprehensive platform developed by Prime Intellect, designed to facilitate the entire lifecycle of agentic model research and optimization. It integrates the Environments Hub with Hosted Training and Hosted Evaluations, enabling researchers and engineers to train, evaluate, and deploy their own models without the need for extensive GPU clusters or intricate algorithmic details.

Unified platform for agentic model research

Integration of Environments Hub, Hosted Training, and Hosted Evaluations

Accessible to researchers and engineers without large-scale infrastructure

Version1.0

RegionUnited States

2026•Prime Intellect

INTELLECT-3 vINTELLECT-3

Paid

AI Models

LLM

A 100B+ parameter Mixture-of-Experts language model trained with large-scale reinforcement learning, achieving state-of-the-art performance on math and reasoning benchmarks. Built on GLM-4.5-Air base with 131K context window.

106B parameter Mixture-of-Experts architecture trained with Prime Intellect RL stack

State-of-the-art math and reasoning performance — evaluated via PRIME-RL and verifiers

$0.20 input / $1.10 output per 1M tokens; 131K token context window with tool use support

VersionINTELLECT-3

RegionUnited States

2026•Prime Intellect

Prime Agent v0.7.0

Prime Agent is an open-source, self-improving coding harness that utilizes Recursive Language Model (RLM) and Continual Harness abstractions to manage complex, long-horizon tasks. It enables agents to programmatically manage their own state, sub-agents, and memory through a persistent IPython kernel, allowing for autonomous, iterative improvement.

Recursive Language Model (RLM) architecture for programmatic sub-agent delegation

Continual Harness for online, reset-free self-improvement of prompts, skills, and memory

Persistent IPython kernel integration for stateful, long-horizon task management

PricingFree (Open-Source)

Version0.7.0