TensorPool
by TensorPool · On-demand multi-node GPU clusters managed from a CLI
BenchRank score
Screenshots of TensorPool
Homepage
Overview
TensorPool provides on-demand multi-node GPU clusters driven from a CLI: 'tp cluster create' provisions nodes with 3.2Tb/s InfiniBand, 'tp storage create' adds shared volumes, and 'tp job push' starts training jobs through a Git-style interface. SSH access is also offered. Capacity is scheduled and binpacked onto clusters TensorPool operates with partners including Nebius, Lambda Labs and Google Cloud.
- Best for
- ML teams training foundation models who want on-demand multi-node GPU clusters driven from a CLI
- Pricing
- No platform fee is shown; GPUs are billed by the hour ($0.015/hr CPU, $1.49/hr L40S, $1.99/hr H100 SXM, $2.99/hr H200 SXM, $4.99/hr B200 SXM, $5.49/hr B300 SXM) with shared storage at $100/TB/month and object storage at $50/TB/month plus $0.005 per 1,000 requests, while Enterprise is contact-sales.
- Runs on
- Web
Strengths and trade-offs
Strengths
- H100 SXM at $1.99/hr versus $3.29 quoted for Lambda Labs
- Multi-node clusters on demand, always 3.2Tb/s InfiniBand
- Shared storage up to 300 GB/s read and 150 GB/s write
- Git-style CLI for training jobs; normal SSH access also offered
Trade-offs
- Standard plan caps at 128 GPUs; beyond that needs Enterprise contact sales
- 99.9% uptime SLA, spend controls and monitoring are Enterprise-only
- Capacity is binpacked onto partner providers, not TensorPool-owned hardware
- Storage is billed separately at $100/TB/month for shared volumes
How TensorPool markets itself
A structured read of the promise, proof and page design on TensorPool’s captured homepage.
Homepage capture
“You focus on the models. We handle the infrastructure.”
- Angle: Autopilot / hands-off
- Hero: Typography only
Pricing
Published plans and prices from TensorPool’s own pricing page.
How this score is made up
Each dimension is scored out of 100 and combined into the headline score using fixed weights.
MCP support
Whether an agent can drive the product through the Model Context Protocol, and how much setup that takes.
Documentation
Publicly reachable docs — coverage, freshness, code samples and machine readability.
Agent friendliness
How readable the site is to an automated client: llms.txt, structured data, server-rendered content, crawler access.
Pricing transparency
Whether real prices are published, self-serve signup exists, and usage costs are knowable without a sales call.
Changelog
A public, dated record of what shipped and when — the clearest signal that a product is still alive.
Marketing site structure
Whether the site answers a buyer's questions: clear positioning, the pages that matter, and accessibility.
Page speed
How fast the site loads for real visitors: Chrome UX Report 75th-percentile LCP, INP and CLS, with a Lighthouse mobile run standing in where a site has too little traffic for field data.
Operational trust
Status page and incident history, security disclosure, compliance and data-processing documentation.
Measured, but not part of the score
Useful to know, but not a mark for or against the product — so these do not affect the ranking.
Openness
Source availability, self-hosting, data export and open standards. Scored and shown, but not part of the composite — paid SaaS is not worse for being paid SaaS.
Maintenance
Release cadence and repository activity. Scored and shown, but not part of the composite — it is only measurable for open repositories.
This doesn’t look right — report a problem with TensorPool’s score
Where this comes from
The TensorPool pages BenchRank reads when it scores the product — its documentation, release notes, status and security pages, and its repository where there is one.
Alternatives in Hosting & Compute
Ranked 1
80.1 — BenchRank score out of 100Cloudflare
Cloudflare · Global network for CDN, security, serverless compute and Zero Trust access
Best for: Teams wanting CDN, DDoS protection, serverless compute and Zero Trust access on one network
Ranked 2
77.2 — BenchRank score out of 100Docker
Docker · Container platform for building, sharing and running applications
Best for: Developer teams that build, share and run container applications and need admin and security controls
Ranked 3
70.6 — BenchRank score out of 100Pulumi
Pulumi · Infrastructure as code in general-purpose programming languages
Best for: Platform teams defining multi-cloud infrastructure in general-purpose programming languages
Is this your product?
Claim TensorPool to manage its profile. Claiming lets you suggest edits to the descriptive fields — it never changes scores or rankings.