Prime Intellect publishes 365,000+ unified RL tasks for SWE, terminal, and search agents
willccbb · x · 2026-07-23
Prime Intellect says it is publishing 365,000+ open agentic RL tasks for SWE, terminal, and search agents, unified behind one API, one sandbox lifecycle, and one command.
- 198k software engineering tasks across 20+ languages
- 28.6k terminal tasks
- 137.6k search tasks
- The release consolidates datasets that previously shipped with different harnesses, image conventions, grading scripts, and failure modes
- It also includes validated and cleaned re-uploads where original datasets needed fixes
The result is a single infrastructure layer for both evals and RL training on Prime Intellect’s platform.
Related event: Prime Intellect Releases 365K Unified Agentic RL Tasks(6 posts)→
More from Infra
- Resin lets an agent build standalone LLM inference engines; vLLM and SGLang optimize DeepSeek kernels differently — AccBalanced · 2026-10-27
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- Anthropic S-1: CSP marketplace revenue jumped from 11% to 48% in three years — AccBalanced · 2026-10-27
- Alchemy adds effect/Workflow support across Durable Objects, Workflows and Lambda Durable Functions — samgoodwin89 · 2026-10-27
- Alchemy's Lambda Web Functions run plain Node web servers on Lambda, no OpenNext needed — samgoodwin89 · 2026-10-27
- Alchemy Cuts Cloudflare Bindings CPU Time by Up to 166× Via Profiling — samgoodwin89 · 2026-10-27