The home of open models, datasets and Spaces.
RunPod
GPU cloud and serverless endpoints for AI.
SharkScore
#31 in Model APIs & inference
Why it scores 62.7
RunPod has a SharkScore of 62.7/100, ranked #31 of 262 in Model APIs & inference (updated 2026-11-18).
Each part compares RunPod with other listings in Model APIs & inference. Evidence covers 87% of the score's weight.
- Adoption
- 41 · 30%
- User reviews
- no data · 25%
- Maintenance
- 100 · 20%
- Trust & transparency
- 65 · 15%
- Momentum
- no data · 10%
Paying never changes this score. How ranking works · method v2026-11-18.2
RankShark review of RunPod
Automated evaluation by RankShark · 18 Nov 2026 · method v1.0 · How we review
We were not paid for this review.
As of 2026-11-18, Runpod offers GPU Pods, Serverless inference and clusters, with listed Pod prices starting at $0.27 per hour.
Runpod is a cloud compute platform for AI and other workloads. It offers GPU and CPU Pods, autoscaling Serverless endpoints, multi-node clusters, storage, and pre-deployed model endpoints. Compute is priced by workload and GPU. [E1, E2, E3]
Best for
- Teams that need GPU compute for model development, training or fine-tuning. · home page, docs
- Teams serving models through API-based, autoscaling inference endpoints. · home page, pricing page, docs
- Teams running multi-GPU or multi-node workloads. · home page, pricing page, docs
Not for
- Teams that need confirmed region-specific data coverage without first checking details with Runpod. · terms
Key features
- GPU Pods. The vendor says Pods provide dedicated GPU instances for development and long-running jobs, with GPU options across more than 30 SKUs. · home page, docs
- Serverless inference. The vendor says Serverless provides API-based GPU endpoints with automatic scaling and usage-based billing. · home page, pricing page
- Clusters. The vendor says Clusters support multi-node GPU workloads; the pricing page lists clusters scaling up to 64 GPUs. · home page, pricing page
- Pre-deployed model endpoints. The vendor says Public Endpoints provide API access to pre-deployed image, video, audio and text models. · pricing page, docs
- Storage. The vendor lists container, volume and network storage, including standard and high-performance network storage. · home page, pricing page
- Global regions. The vendor says workloads can run across 31 global regions; the compliance page says coverage can vary by workload, region, provider and deployment model. · home page, terms
Pros
- The offering covers Pods, Serverless and clusters, with shared access to GPU resources across the product. · home page
- The pricing page publishes GPU-specific rates and storage rates. · pricing page
- The vendor lists pre-deployed model endpoints for several media and text types. · pricing page, docs
- The vendor says it has completed SOC 2 Type II and SOC 3 examinations and maintains HIPAA and GDPR programs. · terms
Cons
- Compute prices vary by GPU and workload; reserved cluster pricing requires contacting sales. · pricing page
- The vendor says compliance coverage can vary by workload, region, provider and deployment model. · terms
- No free plan or free trial stated on the pages we checked.
- Whether customer data is used for model training is not stated on the pages we checked.
Pricing
| Plan | Price | Limits |
|---|---|---|
| Pods | Varies by GPU; listed rates start at $0.27/hr. | Pricing page lists Community Cloud and Secure Cloud. Listed GPU rates include RTX A5000 $0.27/hr; A40 $0.59/hr; RTX A6000 $0.69/hr; RTX 3090 $0.50/hr; A100 PCIe $2.09/hr; H100 PCIe $2.89/hr; B300 $8.99/hr. The page also says billing can be per second. |
| Serverless | Varies by GPU; listed rates start at $0.58/hr. | Listed rates include A4000, A4500, RTX 4000 and RTX 2000 at $0.58/hr; L4, A5000, 3090 and MIG 24GB at $0.75/hr; 4090 at $1.34/hr; H100 at $5.89/hr; B300 at $11.99/hr. The page labels rates per hour and per second. |
| Clusters | Varies by GPU; listed rates start at $2.15/GPU/hr. | Listed rates are A100 SXM $2.15/GPU/hr; H100 SXM $4.79/GPU/hr; H200 SXM $6.35/GPU/hr; B200 $9.59/GPU/hr; B300 $10.79/GPU/hr. The page says clusters scale up to 64 GPUs. |
| Reserved Clusters | Contact sales | The pricing page offers 1mo, 3mo, 6mo, 12mo and 12mo+ terms, and lists B200, H200 SXM, H100 SXM, A100 SXM, B300 and L40S. |
| Storage | Container Disk: $0.10/GB/mo; Volume Disk: Running - $0.10/GB/mo, Idle - $0.20/GB/mo; Network Storage (Standard): Under 1TB - $0.07/GB/mo, Over 1TB - $0.05/GB/mo; Network Storage (High-Performance): $0.14/GB/mo. | Rates are listed per GB per month. |
| Public Endpoints | Prices vary by model and unit; examples include $0.0050 per image request and $0.05 per 1,000 characters. | The pricing page lists per-request, per-character, per-token, per-second and per-megapixel rates for individual audio, image, language and video models. |
Checked 18 Nov 2026 · source
- Free plan
- Unclear: the pages disagree · home page
- Free trial
- Not stated (checked 18 Nov 2026)
- Trains on customer data
- Not stated (checked 18 Nov 2026)
- Data retention
- The privacy policy says personal information is generally retained to fulfill collection purposes, including legal, accounting, reporting, claims and fraud-prevention purposes. It does not give a fixed retention period in the cited text. · privacy page
- Data processing agreement
- The home page says DPAs are available for security review through the Runpod Trust Center; DPA terms are not stated in the evidence. · home page
- AI models used
- Public Endpoints provide access to pre-deployed models. Pricing examples include Whisper V3 Large, FLUX.1 Schnell, Qwen Image, Qwen3 32B AWQ and Seedance 1.0 pro. · pricing page, docs
- API
- The documentation shows API access to Public Endpoints. The release notes say REST API v2 is generally available and that REST API v1 was retired on November 15, 2026. · docs, changelog
- Webhooks
- Not stated (checked 18 Nov 2026)
- Platforms
- Runpod documents connections for Claude Code, Claude Desktop, Cursor, Windsurf and VS Code through its MCP server installation flow. · changelog
- Security claims
- Vendor states that Runpod has completed SOC 2 Type II and SOC 3 examinations. · termsVendor states that its Information Security Management System is certified to ISO/IEC 27001:2022; the page says the certification covers the management system, not individual products or services. · termsVendor states that it maintains HIPAA and GDPR programs. · termsThe vendor home page states 99.9% uptime. · home page
What we measured
- Package downloads · 30 days
- 627,315
- Tranco traffic rank · list of the day
- #17,415
How we tested · sources
- Hourly uptime and response-time checks of the product's URL.
- Repository, release and download data from GitHub, npm, PyPI and Hugging Face where they exist.
- We did not sign up, pay, or use the product as a customer.
- home page: https://www.runpod.io/ · fetched 18 Nov 2026
- pricing page: https://www.runpod.io/pricing · fetched 18 Nov 2026
- docs: https://docs.runpod.io/overview · fetched 18 Nov 2026
- changelog: https://docs.runpod.io/release-notes · fetched 18 Nov 2026
- privacy page: https://www.runpod.io/legal/privacy-policy · fetched 18 Nov 2026
- terms: https://www.runpod.io/legal/compliance · fetched 18 Nov 2026
- security page: https://www.runpod.io/legal/compliance · fetched 18 Nov 2026
Something wrong? Report an error with a public link that shows it; corrections re-run the review.
Top alternatives to RunPod
See all- 78.9ToolFreemium · from $9/mo
- 77.5
API for building with Claude models.
ToolPaid - 76.1
Proxy and SDK to call 100+ LLM APIs in the OpenAI format.
ToolOpen source - 75.7
Google Cloud's platform to build with Gemini and other models.
ToolPaid