Domain on the map · as Enterprise work
RL environments and benchmarks for enterprise workflows
Multi-step business tasks of the kind done inside companies: moving between documents, spreadsheets, internal tools and tickets to finish a job end to end.
- RL environments
- 28
- AI benchmarks
- 139
- Shares entries with
- 16 categories
Snapshot 2026-11-05. How categories are built · Where the data comes from
In short
- What are RL environments for enterprise workflows?
- A simulated workplace where an AI agent practises a task (a code repository, a browser, business software) and an automatic grader scores each attempt. Labs train on thousands of these attempts. Bought to improve models. For enterprise workflows, that means multi-step business tasks of the kind done inside companies: moving between documents, spreadsheets, internal tools and tickets to finish a job end to end.
- Who builds RL environments for enterprise workflows?
- Scale AI, Snorkel AI, Prime Intellect, AfterQuery, Mercor and Turing, among 27 confirmed below.
- What are the main enterprise AI benchmarks?
- Humanity's Last Exam, Terminal-Bench 3.0, Terminal-Bench, OSWorld 2.0, Agents' Last Exam and APEX-Agents, among 93 confirmed below.
RL environments 28
-
Scale AI
Covers enterprise work“Building agentic AI that drives EBITDA gains across PE portfolio companies.” source ↗
- Snorkel AI
-
Prime Intellect
Sells RL environments“Turn any task into an RL environment. Init, develop, eval, and push with the Prime CLI.” source ↗
- AfterQuery
- Mercor
- Turing
- micro1
-
Surge AI
Sells RL environments“We create rich, complex RL environments that challenge agentic models in novel ways, and design verifiers that reward their behavior.” source ↗
-
Fleet AI
Sells RL environments“an applied research lab dedicated to the scale out of environments - diverse and high-fidelity training gyms that accurately model the real world” source ↗
- Deeptune
- Halluminate
- Veris AI
- Collinear
- Huzzle Labs
- Idler
- HUD
-
Aptura
Sells RL environments“Rich environments loaded with real data—verifiable rewards, custom tooling, data licensing support.” source ↗
-
Symbal
Covers enterprise work“a modular AI system that converts unstructured data and manual steps into workflows agents can run.” source ↗
- BenchFlow
-
Plato
Covers enterprise work“A **sim** (simulator) is a pre-built VM image of an application — a CRM, a Git server, a wiki, a Linux desktop.” source ↗
Sells RL environments“Plato is an applied research lab that transforms real-world data into simulated environments for training and evaluating agents.” source ↗
- Ooak Data
-
Theta
Covers enterprise work“Whether it's back office work, sophisticated financial modeling, or refactoring a codebase” source ↗
Sells RL environments“We synthesize your first party data and our human expert data into environments that reflect the real goals, workflows, and outcomes of your business.” source ↗
-
Rise Data Labs
Covers enterprise work“Enterprise SaaS” source ↗
-
AIChamp
Sells RL environments“We build the twin, then robot-AI teams, PLC changes, vision and equipment options are tested in it.” source ↗
-
Habitat Inc
Sells RL environments“Adapt the commit into a task description, golden patch (the solution), and test patch (verification tests).” source ↗
- Metaphi
- Align AI
1 not yet confirmedListed in the catalog; no line from its own site or paper confirms it yet.
AI benchmarks 139
-
Humanity's Last Exam
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Terminal-Bench 3.0
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Terminal-Bench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
OSWorld 2.0
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Agents' Last Exam
Tests enterprise work“a benchmark designed to evaluate AI agents on long horizon, economically valuable, real world tasks with verifiable outcomes.” source ↗
-
APEX-Agents
Tests enterprise work“Long-horizon, cross-application tasks in professional services” source ↗
-
MCP Atlas
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Senior SWE-bench
Snorkel AI · built it with others“Senior SWE-bench built by Snorkel with Princeton University and UW-Madison” source ↗
-
FinanceQA
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
IDE-Bench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Continual Learning Bench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
APEX-Agents 1.1
Tests enterprise work“consultants from McKinsey and BCG” source ↗
-
APEX-SWE
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
APEX-Accounting
Tests enterprise work“Long-horizon, cross-application tasks in professional accounting” source ↗
-
Remote Labor Index
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
HiL-Bench
Scale AI · built it“We design HiL-Bench to reflect this; tasks in this benchmark require agents to ask the right questions for full clarity” source ↗
-
CEO Bench
Tests enterprise work“Executive and departmental decision work inside one simulated company, graded against expert-authored rubrics.” source ↗
-
CWE-bench
Collinear · built it“At Collinear, we built Curator Evals: a benchmarking and evaluation library designed to systematically measure the performance of curators and reward models.” source ↗
-
WebBench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
SpreadsheetBench 2
Tests enterprise work“a workflow-level benchmark for spreadsheet agents that covers three task categories: generation, debugging, and visualization.” source ↗
-
SkillsBench
BenchFlow · built it“We publish the environments and benchmarks we use to evaluate agents on real work — papers, datasets, and live sites. SkillsBench, ClawsBench, Robo Use” source ↗
-
Realm: Pathology-report reasoning benchmark
micro1 · built it“We built Realm legal-reasoning to evaluate legal reasoning as a skill that develops across multiple steps.” source ↗
-
UI-Bench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
App-Bench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Market-Bench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
- YC-Bench
-
VERITAS
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Crosby-micro1 RedlineBench
micro1 · built it with others“Why we built this” source ↗
-
Web Bench (2025-05-30; ~5
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
SWE-Bench Pro v1 (2025-09-19) and v2
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
GDP.xlsx
Tests enterprise work“GDP.xlsx is a benchmark of 70 real-world tasks testing whether frontier models can work with professional Excel workbooks.” source ↗
-
DrugDiscoveryBench with Phylo
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
READY or Not enterprise agent deployment benchmarks
Tests enterprise work“a suite of industry-specific benchmarks that measure agents the way enterprises actually use them: working alongside people, inside real workflows.” source ↗
Scale AI · built it“Today we're introducing READY (Reliable Enterprise Agent Deployment), a suite of industry-specific benchmarks that measure agents the way enterprises actually use them” source ↗
-
RSI Bench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
PRBench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Westworld v1 simulated-web environments (2025-11-23):
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
BrowserBench (2025-11-22):
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
CyberStrike
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
SciCode++
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Open-MM-RL
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
SDLCbench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
CompanyBench
Tests enterprise work“Enterprise financial-analyst agent evaluation on a simulated consumer-fintech neobank.” source ↗
-
EKWbench
Tests enterprise work“Multi-hour operational knowledge work — sourcing reviews, cost roll-ups, assortment resets and referral triage — graded against expert-built answers.” source ↗
-
Medical Bench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
OdysseysMega
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
SWE-Bench Pro - Private Dataset
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Legora BAR
AfterQuery · contributed tasks or data“end-to-end legal tasks drawn from real cases across 28 practice areas, with a public case co-created with AfterQuery.” source ↗
-
CWE-bench v1
Collinear · built it“Evaluations we build to find where models fail” source ↗
- TraitBasis / Impatient Users Confuse AI Agents
-
ClawsBench
Tests enterprise work“Large language model (LLM) agents are increasingly deployed to automate productivity tasks (e.g., email, scheduling, document management)” source ↗
BenchFlow · built it“We publish the environments and benchmarks we use to evaluate agents on real work — papers, datasets, and live sites. SkillsBench, ClawsBench, Robo Use” source ↗
-
Robo Use
BenchFlow · built it“We publish the environments and benchmarks we use to evaluate agents on real work — papers, datasets, and live sites. SkillsBench, ClawsBench, Robo Use” source ↗
-
FrontierPhysics v0.1
BenchFlow · built it“We publish the environments and benchmarks we use to evaluate agents on real work — papers, datasets, and live sites. SkillsBench, ClawsBench, Robo Use, PostTrain, and FrontierPhysics.” source ↗
-
Benchmarking Agents in Insurance Underwriting Environments
Tests enterprise work“to capture real-world enterprise challenges.” source ↗
-
BigLaw Bench: Research
Snorkel AI · contributed tasks or data“Snorkel AI served as Harvey’s data partner for BigLaw Bench: Research, bringing its expertise in creating complex expert data for frontier AI” source ↗
-
Measuring Autonomous AI Research
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
General Agent synthetic environment
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Realm: Legal reasoning benchmark
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Realm: Financial reasoning benchmark
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
DAYJOB (Healthcare/Finance) - arXiv:2610.01306
Surge AI · built it“We introduce DAYJOB, 130 long-horizon tasks built by professionals in healthcare and finance that start from a brief request” source ↗
-
ComplexConstraints and Beyond
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
EnterpriseBench CoreCraft - arXiv:2602.16179
Tests enterprise work“CoreCraft is a fully operational enterprise simulation of a customer support organization, comprising over 2,500 entities across 14 entity types with 23 unique tools” source ↗
Surge AI · built it“We introduce CoreCraft, the first environment in EnterpriseBench, Surge AI's suite of agentic RL environments.” source ↗
-
The Hierarchy of Agentic Capabilities - arXiv:2601.09032
Tests enterprise work“We present an empirical study evaluating frontier AI models on 150 workplace tasks within a realistic e-commerce RL environment from Surge.” source ↗
-
Antidote and Hemingway-bench leaderboards
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Westworld Finance Diligence Bench (2026-08-11):
Tests enterprise work“an agent may need to search a data room, reconcile spreadsheet models, draft client-facing documents, and communicate results over email” source ↗
-
DealTrace Bench (2026-05/06):
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
VAmoS Pro Bench / 'VAmoS Part Deux'
Tests enterprise work“a benchmark that combines these challenges in 100 calls about utility billing and payment assistance” source ↗
-
VAmoS Bench: Voice Agent Simulation Bench
Tests enterprise work“It measures complete voice-agent systems end to end in a stateful customer-support task.” source ↗
- Public leaderboards VAmoS and VAmoS Pro
-
BankingBench
Tests enterprise work“Each benchmark session is a full simulated support interaction: a customer contacts the agent about a card issue” source ↗
Veris AI · built it“Veris AI is a simulation platform for AI agent development, and we ran 7 versions of our customer service agent through 100 card-replacement scenarios.” source ↗
-
SantaBench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
VADER
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
LibraryDesignBench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
SlopCodeBench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
APEX (AI Productivity Index) v1 / APEX-1
Tests enterprise work“Tests whether frontier models can perform economically valuable tasks across professional domains including investment banking, corporate law, management consulting, and medicine.” source ↗
-
SWE-Marathon-Ext
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Realm: Tax reasoning benchmark
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
CortexRetrievalBench
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
LongExtractionBench
Tests enterprise work“Much of the highest-value work in enterprise data extraction lives in long, content-rich, table-heavy documents” source ↗
-
Chartography - arXiv:2608.10677
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
HANDBOOK.md - arXiv:2607.25398
Tests enterprise work“instructs it to carry out routine professional work governed by an expert-written standard operating procedure of 20-124 pages” source ↗
-
GDP.pdf - arXiv:2607.11192
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Riemann-Bench - arXiv:2604.06802
Surge AI · built it“We introduce RIEMANN-BENCH, a private benchmark of 25 expert-curated problems designed to evaluate AI systems on research-level mathematics” source ↗
-
Assemble Bench (robotics
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
SheetBench-50 (with Sepal AI) - 2025-10-01
Tests enterprise work“Calculate the # of unique customer IDs in the worksheet ANSWER cell A1 and calculate the # of duplicate IDs in cell A2.” source ↗
-
InsureBench: insurance underwriting and claims benchmark
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
CorpLaw (measured 2026-08-18; access on request):
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
AdvancedIF - arXiv
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
- RFT case study with Savi Security, Qwen3 4B scam detection
-
HANDBOOK.md Agents
Tests enterprise work“Each task is a unique RL environment with internal tools and external MCP servers, across five enterprise domains.” source ↗
-
SWE Atlas: Codebase QnA, Test Writing, Refactoring
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
-
Tuesday Work Index leaderboard
No line from its own site or paper places it in Enterprise Workflows yet; the catalog lists it here.
- ShelfLife (measured 2026-08-13):
-
ShelfLife E-Sim (measured 2026-08-26):
Tests enterprise work“E-Sim is a long-horizon resource management benchmark built on a stateful simulator of retail operations.” source ↗
46 not yet confirmedListed in the catalog; no line from its own site or paper confirms it yet.
- CREW
- Turing Terminal-Bench Data
- OSWorld-Verified
- Technical Report
- Agent-environment papers
- KernelQuest
- PLSQLBench
- Turing Frontier Agentic STEM
- The Valley of Code Reasoning
- Cats Confuse Reasoning LLM
- Self-rationalization improves LLM as a fine-grained judge
- env0 SFT study on SkillsBench/ClawsBench
- GenesisBench
- FinQA / SnorkelFinance environment and benchmark
- RIFT / Hitchhiker's Guide to Rubric Quality
- Shrinking the Generation-Verification Gap with Weak Verifiers
- Snorkel: Rapid Training Data Creation with Weak Supervision
- Human Baselines for Benchmarks
- INTELLECT-3 Technical Report
- INTELLECT-2
- Systematic Reward Hacking study
- True Agents Model the World / ECHO
- Scaling Agentic RL: 365k+ SWE/terminal/search tasks
- No Last Mile: A Theory of the Human Data Market
- PII Transformation for Enterprise Datasets
- AI Models Now Introduce Safety Risks in the Physical World
- Cross-Benchmark Transfer from RL on Agentic Coding Tasks - arXiv
- Post-Training on Office Work Improves Software Engineering - arXiv
- Cross-Benchmark Generalization in Long-Horizon Agents - arXiv
- Simulated-to-Real Transfer for Web Agents (2026-07-27):
- ApexAgents SkyRL 397B RL training recipe
- The Benchmark Ceiling
- LiveWeb samples (with Browserbase) - 2026-09-22/28
- Production-debugging RL environment (hud-ops-diagnostics) - 2026-01-20
- OpenEnv Visual Memory and Spreadsheet environments on Hugging Face
- User-simulation realism: Veris vs τ²-bench
- Printing Machines (2026-01-28)
- No peer-reviewed paper authored by Fleet found. Harbor
- HuzzleWorld computer-use models
- SheetBench-50 (released 2025-10-01)
- Scale Labs (launched 2026-03-09
- PostTrain Arena
- Learning from Less: RLVR in Low Data Regimes
- ACE (AI Consumer Index)
- Universal offline sandbox escape disclosure
- Rubric-based RL papers
Data 23
- AfterQuery
- Align AI
- Aviro
-
Chakra Labs
Sells data“Deterministic environments with frame-accurate state control, high-fidelity trajectory datasets, and mixed-modality training capability.” source ↗
Covers enterprise work“An open platform for interactive clones of production software built to the same standard as frontier lab infrastructure.” source ↗
- Collinear
- Cua
-
Datacurve
Sells data“Prebuilt datasets, curated for signal, reviewed for quality, and structured to drop into your training stack without translation work.” source ↗
Covers enterprise work“From enterprise data systems with layered schemas, to enterprise software systems with evolving business logic and real integration constraints” source ↗
-
Handshake
Sells data“Exploring why human judgment, skill, and expertise remain essential to training modern AI and how people shape the future of artificial intelligence.” source ↗
Covers enterprise work“a workbook whose formulas must tie across tabs, a deck whose charts must match the underlying data, or an email that must actually have been sent.” source ↗
- HUD
- Huzzle Labs
- Markov
- Mercor
- micro1
- Patronus AI
-
ReasonCore
Sells data“We produce exclusive and off-the-shelf (OTS) datasets built around multi-step hierarchical reasoning, tool-augmented workflows, and domain-specific problems” source ↗
Covers enterprise work“These environments are modeled on enterprise software at its most demanding — operational dashboards, financial platforms, infrastructure tooling, compliance workflows” source ↗
- Refresh
- Rise Data Labs
- Scale AI
- Snorkel AI
- Symbal
-
Theta
Covers enterprise work“Whether it's back office work, sophisticated financial modeling, or refactoring a codebase” source ↗
-
Toloka
Sells data“Our vast network of vetted coding experts offers scalable data production for the languages, coding domains, and programming expertise of your choice.” source ↗
Covers enterprise work“Model Context Protocol replicas of enterprise tools with realistic schemas, data flows, and permission models.” source ↗
- Turing
Next: pick another domain. Tags come from the RL Research daily scrape, canonicalized nightly.