BenchFlow C Verified
RL environment company · Enterprise Workflows
- On X
- @benchflow_ai ↗
- Tier
- C why
- Snapshot
What BenchFlow does
BenchFlow is an open-source frontier environment lab that builds RL environments and agent benchmarks (SkillsBench, ClawsBench, Robo Use, FrontierPhysics) plus a runtime for running them.
- Sells
- RL environments and Infrastructure
- Leads with
- Several domains (a generalist)
- Based in
- San Francisco, USA
- Team
- 1-10
- Website
- benchflow.ai ↗
What BenchFlow sells
Read from BenchFlow's own website, checked October 11, 2026: the line that puts it in each category.
RL environments
“BenchFlow builds the environments AI agents learn in”
benchflow.ai/ ↗“Do procedural skills make agents better at real work? 86 tasks, 11 domains.”
benchflow.ai/ ↗
Infrastructure
“Run a real task locally: Docker, an ACP agent, an independent verifier.”
benchflow.ai/ ↗“Contribute an environment. We post-train and score what generalizes.”
benchflow.ai/ ↗
Domains BenchFlow covers
“the internal-dogfood stateful multi-service benchmark (mock Gmail / Slack / Calendar / Docs / Drive)”
benchflow.ai/docs/benchflow/environment-plane ↗“Coding agents drive simulated robots from the shell. 341 tasks, 14 suites.”
benchflow.ai/ ↗“Evaluating agents for end-to-end frontier physics research.”
benchflow.ai/ ↗“Five wire-compatible workplaces. Capability and safety, scored separately.”
benchflow.ai/ ↗
Benchmarks BenchFlow built or runs
ClawsBench 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”
benchflow.ai/research ↗FrontierPhysics v0.1 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.”
benchflow.ai/research ↗PostTrain Arena ↗ 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.”
benchflow.ai/research ↗Robo Use 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”
benchflow.ai/research ↗SkillsBench 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”
benchflow.ai/research ↗
BenchFlow competitors and alternatives
The companies that sell the most of the same things in the same domains, and what they share with BenchFlow.
- ReasonCore Also sells RL environments and evals · also in robotics, science and enterprise work
- HUD Also sells RL environments, training platforms, evals and sandboxes and compute · also in enterprise work
- Prime Intellect Also sells RL environments, evals, training platforms and sandboxes and compute · also in science
- Collinear Also sells RL environments and evals · also in security and enterprise work
- Mercor Also sells evals and RL environments · also in enterprise work and science
Tier C
Early: a small team.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- team 1-10
- raised $1M
- investors: Founders, Inc., Y Combinator, Pear VC
- in curated RL directories (5 pts)
On X · unclear, low buzz, from 1 post about BenchFlow in the last month
Only mention is a low-engagement post about BenchShield, a benchmark-vulnerability research collaboration with Dartmouth and Berkeley, saying little about BenchFlow's own products.
How we got to tier C
- Adoption: team 1-10; raised $1M; investors: Founders, Inc., Y Combinator, Pear VC. Tier C.
- On X: unclear, low buzz. “Only mention is a low-engagement post about BenchShield, a benchmark-vulnerability research collaboration with Dartmouth and Berkeley, saying little about BenchFlow's own products.” No move.
- In curated RL directories: 5 points, a floor of C.
- Confirmed: a line from its own site or paper places it.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Where BenchFlow sits
Its domain on the map, live. Hover a logo for its name; click to open it. Open it on the full map
Latest from BenchFlow on X
@benchflow_ai ↗Next: browse all environments. Entry from the RL Research daily scrape of public sources, 2026-11-05 snapshot.