# EnvIndex artifact catalog As of 2026-08-16. Unknown means no reliable evidence is recorded; it does not mean No. Private describes access, not verification. Canonical methodology: https://envindex.com/methodology ## RL Environments and Agents - Canonical: https://envindex.com/environments/surge-rl-environments-and-agents - Vendor: Surge - Type: RL Environment - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: enterprise, multi-domain, long-horizon, tool-use - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Custom RL environments and verifier design for training and evaluating agentic models. - Sources: https://surgehq.ai/products ## Rubrics and Verifiers - Canonical: https://envindex.com/environments/surge-rubrics-and-verifiers - Vendor: Surge - Type: Verifier - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: alignment, multi-domain - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Custom rubric and verifier design for scoring complex model and agent behavior. - Sources: https://surgehq.ai/products ## RLHF - Canonical: https://envindex.com/environments/surge-rlhf - Vendor: Surge - Type: Dataset - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: rlhf, alignment, multi-domain - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Human preference and reward data for reinforcement learning from human feedback. - Sources: https://surgehq.ai/products ## Supervised Fine-Tuning Data - Canonical: https://envindex.com/environments/surge-sft - Vendor: Surge - Type: Dataset - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: computer-use, browser, multi-domain - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Expert demonstrations for bootstrapping model capabilities, including computer and browser use. - Sources: https://surgehq.ai/products ## Human Evaluation - Canonical: https://envindex.com/environments/surge-human-evaluation - Vendor: Surge - Type: Evaluation Suite - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: alignment, multi-domain - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Human evaluation programs for model quality, usefulness, safety, and subjective output characteristics. - Sources: https://surgehq.ai/products ## Expert Professional Domains - Canonical: https://envindex.com/environments/surge-expert-professional-domains - Vendor: Surge - Type: Dataset - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: finance, legal, medical, science, math, multi-domain - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Expert-authored data and judgment across professional, STEM, and humanities domains. - Sources: https://surgehq.ai/products ## Internationalization - Canonical: https://envindex.com/environments/surge-internationalization - Vendor: Surge - Type: Dataset - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: multi-domain - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Language and culturally grounded training data across more than 70 reported languages. - Sources: https://surgehq.ai/products ## Multimodal Data - Canonical: https://envindex.com/environments/surge-multimodal-data - Vendor: Surge - Type: Dataset - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: multi-domain - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Training and evaluation data spanning text, images, audio, and video. - Sources: https://surgehq.ai/products ## Off-The-Shelf Data Catalog - Canonical: https://envindex.com/environments/surge-off-the-shelf-data - Vendor: Surge - Type: Environment Bundle - Availability: private-commercial - Access: Private catalog, Commercial, Off-the-shelf - Domains: coding, enterprise, science, math, tool-use, multi-domain, long-horizon - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: Pre-built commercial datasets and RL environments spanning coding, enterprise agents, STEM, tool use, and reasoning. - Sources: https://surgehq.ai/ots ## CoreCraft - Canonical: https://envindex.com/environments/surge-corecraft - Vendor: Surge - Type: Benchmark - Availability: public-sample - Access: Public / Open - Domains: enterprise, tool-use, long-horizon - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: A simulated enterprise environment where agents complete customer-support and operational tasks inside a fictional PC retailer. - Sources: https://surgehq.ai/blog/rl-envs-real-world, https://surgehq.ai/benchmarks ## HANDBOOK.md Agents - Canonical: https://envindex.com/environments/surge-handbook-agents - Vendor: Surge - Type: Benchmark - Availability: public-sample - Access: Public / Open - Domains: enterprise, tool-use, long-horizon - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: A long-context enterprise-agent benchmark built from unique RL environments with internal tools and external MCP servers. - Sources: https://surgehq.ai/benchmarks ## Long-Horizon Knowledge-Work Tasks - Canonical: https://envindex.com/environments/rise-data-labs-long-horizon-knowledge-work-tasks - Vendor: Rise Data Labs - Type: Environment Bundle - Availability: private-commercial - Access: Private catalog, Commercial - Domains: enterprise, finance, legal, science, multi-domain, long-horizon - Verification: vendor-submitted (medium confidence), last verified 2026-08-16 - Summary: Private commercial long-horizon tasks spanning multiple knowledge-work domains. - Sources: https://risedatalabs.com/solutions/rl-environment ## Finance Tasks - Canonical: https://envindex.com/environments/rise-data-labs-finance-tasks - Vendor: Rise Data Labs - Type: Environment Bundle - Availability: private-commercial - Access: Private catalog, Commercial - Domains: finance, enterprise - Verification: vendor-submitted (medium confidence), last verified 2026-08-16 - Summary: A private commercial catalog of finance tasks for RL training and evaluation. - Sources: https://risedatalabs.com/solutions/rl-environment ## Custom Enterprise Environments - Canonical: https://envindex.com/environments/rise-data-labs-custom-enterprise-environments - Vendor: Rise Data Labs - Type: RL Environment - Availability: custom-capability - Access: Custom-built, Commercial, Request demo - Domains: enterprise, finance, legal, cybersecurity, multi-domain - Verification: company-reported (medium confidence), last verified 2026-08-16 - Summary: A custom environment-development capability based on a buyer’s workflows, tools, data, and success criteria. - Sources: https://risedatalabs.com/solutions/rl-environment ## GBA Eval - Canonical: https://envindex.com/environments/gba-eval - Vendor: Mechanize - Type: Benchmark - Availability: public-sample - Access: Public / Open - Domains: coding, long-horizon - Verification: company-reported (high confidence), last verified 2026-08-16 - Summary: A long-horizon coding eval in which agents build a Game Boy Advance emulator in Rust within 24 hours. - Sources: https://www.mechanize.work/technical-blog/introducing-gba-eval/, https://gbaeval.com ## SWE-Bench Pro - Canonical: https://envindex.com/environments/swe-bench-pro - Vendor: Scale - Type: Benchmark - Availability: public-sample - Access: Public / Open, Open source - Domains: coding, long-horizon - Verification: confirmed (high confidence), last verified 2026-08-16 - Summary: A long-horizon software-engineering benchmark built from public and proprietary repositories. - Sources: https://labs.scale.com/papers/swe_bench_pro, https://labs.scale.com/leaderboard/swe_bench_pro_public ## SWE-Bench Pro - Private Dataset - Canonical: https://envindex.com/environments/swe-bench-pro-private - Vendor: Scale - Type: Benchmark - Availability: private-commercial - Access: Private catalog, Enterprise-only - Domains: coding, long-horizon, enterprise - Verification: company-reported (medium confidence), last verified 2026-08-16 - Summary: The proprietary-repository subset of SWE-Bench Pro, documented publicly but not distributed as an open dataset. - Sources: https://labs.scale.com/leaderboard/swe_bench_pro_private ## Gandalf the Grader - Canonical: https://envindex.com/environments/gandalf-the-grader - Vendor: Handshake - Type: Verifier - Availability: public-sample - Access: Public / Open, Open source - Domains: finance, enterprise, tool-use - Verification: confirmed (high confidence), last verified 2026-08-16 - Summary: An open-source reactive agent-as-judge that inspects files and environment state while grading agent work. - Sources: https://joinhandshake.com/research/ai/gandalf-the-grader/, https://github.com/Handshake-AI-Research/gandalf-the-grader ## BankerToolBench - Canonical: https://envindex.com/environments/bankertoolbench - Vendor: Handshake - Type: Benchmark - Availability: public-sample - Access: Public / Open - Domains: finance, enterprise, tool-use - Verification: confirmed (medium confidence), last verified 2026-08-16 - Summary: A banking-workflow benchmark used to evaluate agents and verifier behavior on artifact-heavy tasks. - Sources: https://huggingface.co/datasets/handshake-ai-research/bankertoolbench ## App-Bench - Canonical: https://envindex.com/environments/app-bench - Vendor: AfterQuery - Type: Benchmark - Availability: public-sample - Access: Public / Open - Domains: coding, design - Verification: company-reported (medium confidence), last verified 2026-08-16 - Summary: A benchmark for evaluating AI systems that generate modern web applications. - Sources: https://www.afterquery.com/leaderboard/app-bench ## APEX - Canonical: https://envindex.com/environments/apex - Vendor: Mercor - Type: Benchmark - Availability: public-sample - Access: Public / Open - Domains: finance, legal, medical, enterprise - Verification: confirmed (high confidence), last verified 2026-08-16 - Summary: An expert-authored benchmark of economically valuable tasks in finance, consulting, law, and medicine. - Sources: https://www.mercor.com/research/, https://huggingface.co/datasets/Mercor/APEX