# Cerebrium

Real-time AI infrastructure that scales with you

Source: https://wheredoirunthis.com/products/cerebrium · JSON: https://wheredoirunthis.com/api/products/cerebrium · Read on 2026-10-14

## At a glance

- Type: Inference cloud
- Kind: Hosted
- Isolation: Kata / gVisor
- Free tier: Yes
- Starting price: $100/mo
- GPU: Yes
- Regions: 4
- Coverage: 18 of 61 assessed capabilities supported
- Agent readiness: read 11 of 11 signals, drive 9 of 11 signals

## Links

- Homepage: https://cerebrium.ai/
- Docs: https://cerebrium.ai/docs/getting-started/introduction
- Pricing: https://cerebrium.ai/pricing
- Repository: https://github.com/CerebriumAI/cerebrium

## Pricing

- Model: Per-second compute (GPU + vCPU + memory) + plan tier (concurrency/seats); storage GB-month
- Unit: GPU-second (by type) + vCPU-second + GB-RAM-second; storage GB-month
- Free tier: Yes

| Tier | Price | Includes | Notes |
| --- | --- | --- | --- |
| Hobby | Free + compute | 3 seats; up to 3 apps; 500 containers + 5 concurrent GPUs; logs conflict (1 vs 7 days); first 100 GB storage free.; Free plan; pay only for per-second compute used; 5 GPU concurrency. |  [source](https://cerebrium.ai/pricing) |
| Standard | $100/mo + compute | Unlimited apps; 1000 containers + 30 concurrent GPUs; custom domains; 30-day logs; seats conflict (unlimited vs 10).; Same per-second compute rates as Hobby; higher concurrency (30 GPU); custom domains. |  [source](https://cerebrium.ai/pricing) |
| Enterprise | Custom | unlimited GPU concurrency · dedicated Slack · white-glove onboarding · ML engineering services · unlimited log retention · HIPAA/GDPR/ISO 27001 | Contact sales. [source](https://cerebrium.ai/pricing) |

- Overages: Per-second GPU $0.000164/s (T4) → $0.00167/s (B200); CPU $0.00000655/vCPU/s; memory $0.00000222/GB/s; storage $0.05/GB-mo (100 GB free)
- Card required: Not documented
- Source: https://cerebrium.ai/pricing

## Website design

Desktop public website · assessed 2026-10-14

Complete assessment.

Dark purple/pink expressive marketing contrasts with pale structured pricing and a light three-column documentation surface. Desktop viewport only; consent overlays partly obscure marketing copy.

- Styles: Corporate
- Landing page: captured
- Pricing: captured
- Docs: captured

## Runtime

- **GPU & accelerators**: Yes — "B200, H200, H100, RTX PRO 6000, A100, L40s, A10, L4 and T4 per-second rates." [source](https://cerebrium.ai/pricing) (confidence: high)
- **BYOC (Bring your own cloud)**: No — "No customer-cloud deployment documented on the reviewed hosted-platform surface." [source](https://cerebrium.ai/docs/llms.txt) (confidence: medium)
- **Secrets management**: Encrypted project/app secrets exposed as environment variables; API and CLI management. [source](https://cerebrium.ai/docs/other-topics/using-secrets.md)
- **Storage options**: Persistent project volumes per region; shared global volume for global apps; first 100 GB storage free, then $0.05/GB-month. [source](https://cerebrium.ai/docs/storage/managing-files.md)
- **Supported workloads**: real-time-inference, voice-agents, llm-inference, video-models, custom-python-servers, batch-training — "REST/SSE/WebSockets and ASGI apps; homepage training example; primary inference positioning." [source](https://cerebrium.ai/docs/getting-started/introduction.md) (confidence: high)
- **Persistence**: Region-local /persistent-storage; global apps additionally mount /global-persistent-storage. Files are not automatically synchronized between regional volumes. [source](https://cerebrium.ai/docs/deployments/multi-region-deployment.md)
- **Data jurisdiction**: AWS and Nebius regions in US and Europe; additional regions available on request. Placement supports global, US/EU zones or specific region. [source](https://cerebrium.ai/docs/deployments/multi-region-deployment.md)
- **Resource limits & quotas**: Hobby: 3 apps, 3 seats, 500 containers / 5 concurrent GPUs. Standard: unlimited apps, 1000 containers / 30 GPUs. Pricing contradicts itself on Standard seats (unlimited vs 10) and Hobby logs (1 vs 7 days). [source](https://cerebrium.ai/pricing)
- **SDK languages**: Not documented — "Current quickstart installs the CLI and executes ordinary Python; no separate official client SDK established from current docs." [source](https://cerebrium.ai/docs/getting-started/introduction.md) (confidence: medium)
- **Network & egress controls**: No — "No outbound policy/firewall primitive found; private inter-cluster routing is a separate capability." [source](https://cerebrium.ai/docs/llms.txt) (confidence: medium)
- **Snapshots & forks**: Yes — "Memory and GPU Checkpointing (Beta); restores CPU/GPU state to reduce cold starts, not general VM forking." [source](https://cerebrium.ai/docs/performance/checkpointing.md) (confidence: high)
- **Deployment methods**: cli, api, dockerfile, container-image, git-push — "Official CLI, custom Dockerfiles/private base images, REST control API and GitHub-linked deployments." [source](https://cerebrium.ai/docs/llms.txt) (confidence: high)
- **Isolation**: Kata / gVisor — "We run each workload on top of gVisor in a hardened, isolated environment to provide strong container isolation without compromising performance." [source](https://cerebrium.ai/) (confidence: high)

## Capabilities

### AI & agents

| Capability | Status | Evidence |
| --- | --- | --- |
| Agent skills | Supported | "name: cerebrium" [source](https://cerebrium.ai/docs/skill.md) |
| AI app builder | Not supported | "No advertised app-builder capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Autonomous QA agent | Not supported | "No advertised autonomous-qa-agent capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Built-in coding agent | Not supported | "No advertised built-in-agent capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| BYOK for LLMs | Not supported | "No advertised byok-llm capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Credential-injection proxy | Not supported | "No advertised credential-injection-proxy capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| LLM gateway | Not supported | "No advertised llm-gateway capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| MCP client | Not supported | "No advertised mcp-client capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| MCP-server hosting | Not supported | "No advertised mcp-hosting capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| MCP server | Supported | "This Model Context Protocol server provides search and retrieval tools for the Cerebrium site." [source](https://cerebrium.ai/docs/mcp) |
| Model inference API (first-party) | Supported | "Build an OpenAI compatible chat completions endpoint on Cerebrium" [source](https://cerebrium.ai/docs/llms.txt) |
| Preinstalled coding agent | Not supported | "No advertised preinstalled-coding-agent capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |

### Networking & delivery

| Capability | Status | Evidence |
| --- | --- | --- |
| Automatic HTTPS | Supported | "https://api.cerebrium.ai/v4/p-xxxxxxxx/{app-name}/{function}" [source](https://cerebrium.ai/docs/endpoints/inference-api.md) |
| Custom domains | Supported | "Custom domains" [source](https://cerebrium.ai/pricing) |
| Direct public IPv6 | Not supported | "No advertised direct-ipv6 capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Edge KV store | Not supported | "No advertised edge-kv capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Egress controls | Not supported | "No advertised egress-controls capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Email send / receive | Not supported | "No advertised email-send-receive capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Public preview URLs | Partially supported | "Once deployed, the app is callable at a POST endpoint:" [source](https://cerebrium.ai/docs/getting-started/introduction.md) |
| Private networking | Supported | "connect Cerebrium apps privately within a region" [source](https://cerebrium.ai/docs/networking/inter-cluster-routing.md) |
| Built-in reverse proxy | Supported | "Requests enter through the global router and are directed to a region where the app is running." [source](https://cerebrium.ai/docs/deployments/multi-region-deployment.md) |

### Organisation & security

| Capability | Status | Evidence |
| --- | --- | --- |
| Built-in auth | Partially supported | "Authenticate using the JWT token from the **API Keys** section of the dashboard." [source](https://cerebrium.ai/docs/endpoints/inference-api.md) |
| BYOC | Not supported | "No advertised byoc capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Deploy into customers' VPCs | Not supported | "No advertised customer-vpc-deploy capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Secrets management | Supported | "Store API keys and credentials as encrypted secrets in Cerebrium" [source](https://cerebrium.ai/docs/other-topics/using-secrets.md) |
| Self-host / OSS core | Not supported | "No advertised self-host-oss capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Sharing | Not supported | "No advertised sharing capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| SSO / SAML | Not supported | "No advertised sso-saml capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Teams | Supported | "3 user seats" [source](https://cerebrium.ai/pricing) |
| Workload identity | Not supported | "Service Account Token authentication." [source](https://cerebrium.ai/docs/api-reference/api-keys/create-api-key.md) |

### Developer access

| Capability | Status | Evidence |
| --- | --- | --- |
| CLI | Supported | "pip install cerebrium" [source](https://cerebrium.ai/docs/getting-started/introduction.md) |
| E2B-compatible API | Not supported | "No advertised e2b-compatible-api capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Filesystem API | Partially supported | "Download a file from a specific volume." [source](https://cerebrium.ai/docs/llms.txt) |
| Git deploy | Supported | "CI/CD Pipelines" [source](https://cerebrium.ai/docs/deployments/ci-cd.md) |
| HTTP data API | Not supported | "No advertised http-query-api capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| OpenAPI spec | Supported | ""openapi": "3.0.0"" [source](https://s3.eu-west-1.amazonaws.com/www.cerebrium.ai/openapi_spec.json) |
| Private container registry | Not supported | "Authenticate with Docker Hub, AWS ECR, or other private registries" [source](https://cerebrium.ai/docs/container-images/private-docker-registry.md) |
| Public API | Supported | "REST API for interacting with Cerebrium." [source](https://cerebrium.ai/docs/api-reference/api-keys/create-api-key.md) |
| SDK | Not supported | "pip install cerebrium" [source](https://cerebrium.ai/docs/getting-started/introduction.md) |
| SSH access | Not supported | "No advertised ssh-access capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |

### Compute & storage

| Capability | Status | Evidence |
| --- | --- | --- |
| Cloud-init / ignition config | Not supported | "No advertised cloud-init-config capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Desktop / computer-use sandbox | Not supported | "No advertised desktop-sandbox capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| GPU compute | Supported | "B200; H200; H100; RTX PRO 6000; A100 (80GB); A100 (40GB); L40s; A10; L4; T4." [source](https://cerebrium.ai/pricing) |
| Live-state (memory) snapshots | Partially supported | "Snapshot CPU and GPU memory to skip imports, model loading, and CUDA kernel compilation" [source](https://cerebrium.ai/docs/performance/checkpointing.md) |
| Multi-region | Partially supported | "Multi-region deployment is currently in **beta**." [source](https://cerebrium.ai/docs/deployments/multi-region-deployment.md) |
| Object storage | Not supported | "No advertised object-storage capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Persistent disk | Supported | "Cerebrium offers file management through a persistent volume that's available to all apps in a project." [source](https://cerebrium.ai/docs/storage/managing-files.md) |
| Sandbox templates / custom images | Supported | "Deploy containerized apps on Cerebrium with your own Dockerfile" [source](https://cerebrium.ai/docs/llms.txt) |
| Scale to zero | Supported | "with scale-to-zero and per-second billing." [source](https://cerebrium.ai/docs/skill.md) |
| Selectable isolation runtime | Not supported | "We run each workload on top of gVisor" [source](https://cerebrium.ai/) |
| Snapshots / forking | Partially supported | "Memory and GPU Checkpointing (Beta)" [source](https://cerebrium.ai/docs/performance/checkpointing.md) |

### Data

| Capability | Status | Evidence |
| --- | --- | --- |
| Database branching | Not supported | "No advertised database-branching capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Ephemeral instant DB (no-signup) | Not supported | "No advertised ephemeral-instant-db capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| First-party database | Not supported | "No advertised first-party-db capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Vector search | Not supported | "No advertised vector-search capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |

### Platform & operations

| Capability | Status | Evidence |
| --- | --- | --- |
| Managed CI/CD runners | Not supported | "No advertised managed-ci-runners capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Built-in observability | Supported | "Logs, metrics, scaling events, and system performance, all in real time." [source](https://cerebrium.ai/) |
| Preview environments | Not supported | "No advertised preview-environments capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Scheduled tasks | Not supported | "No advertised scheduled-tasks capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Project template marketplace | Not supported | "No advertised template-marketplace capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |
| Durable workflows | Not supported | "No advertised workflows-durable capability found in the current documentation index, landing page or pricing surface." [source](https://cerebrium.ai/docs/llms.txt) |

## Agent readiness

### Read: 11 of 11 signals present

11/11 signals (9 full, 2 partial)

- Llms txt: pass — "https://cerebrium.ai/docs/llms.txt — Substantive docs and apex indexes with operational pages and pricing links."
- Llms full txt: pass — "https://cerebrium.ai/docs/llms-full.txt — 405925-byte comprehensive Markdown document fetched; apex variant 404 but docs copy exists."
- Robots AI policy: pass — "https://cerebrium.ai/robots.txt — Explicit AI crawler allow rules."
- JSON LD: pass — "https://cerebrium.ai/ — Organization, WebSite and SoftwareApplication with offers."
- Sitemap: pass — "https://cerebrium.ai/sitemap.xml — Nontrivial XML urlset."
- Meta quality: pass — "https://cerebrium.ai/ — Descriptive title, description, OG and Twitter metadata."
- Markdown negotiation: partial — "https://cerebrium.ai/docs/getting-started/introduction.md — Homepage Accept: text/markdown still HTML; documentation .md URLs serve clean Markdown."
- Content efficiency: partial — "https://cerebrium.ai/ — 7985 extracted readable-text bytes / 472937 HTML bytes, between 1% and 5%."
- Docs discoverable: pass — "https://cerebrium.ai/ — Docs link in first-class navigation."
- Pricing machine readable: pass — "https://cerebrium.ai/pricing — Parseable GPU/CPU/memory/storage rates and tiers; two plan-label inconsistencies explicitly recorded."
- Anti bot: pass — "https://cerebrium.ai/ — Public landing/pricing/docs and evidence fetched without challenge."

### Drive: 9 of 11 signals present

9/11 signals (8 full, 1 partial)

- Public API: pass — "https://cerebrium.ai/docs/api-reference/api-keys/create-api-key.md — Documented REST control API and inference endpoints."
- Openapi spec: pass — "https://s3.eu-west-1.amazonaws.com/www.cerebrium.ai/openapi_spec.json — Valid OpenAPI 3.0.0 JSON; 366688 bytes fetched from docs-linked S3 URL."
- MCP server: pass — "https://cerebrium.ai/docs/mcp — Documented HTTP MCP endpoint; documentation retrieval only, not workload control."
- SDK packages: fail — "https://cerebrium.ai/docs/getting-started/introduction.md — Current documented Python package installs CLI; no separate client SDK verified."
- Auth docs: pass — "https://cerebrium.ai/docs/api-reference/api-keys/create-api-key.md — Bearer service-account token creation, rotation and expiry documented."
- Agent self serve creds: pass — "https://cerebrium.ai/docs/api-reference/api-keys/create-api-key.md — Programmatic API key creation documented; bootstrapping service account still requires dashboard authentication."
- Webhooks: pass — "https://cerebrium.ai/docs/endpoints/webhook.md — Documented response forwarding, retries and HMAC signature verification."
- CLI tool: pass — "https://cerebrium.ai/docs/getting-started/introduction.md — Official pip/brew/Linux/Windows CLI installation."
- Error model: pass — "https://cerebrium.ai/docs/api-reference/api-keys/create-api-key.md — 400 and 401 JSON schema has message string with documented meanings."
- Rate limit docs: fail — "https://cerebrium.ai/docs/llms.txt — Concurrency caps documented, but no request-rate response-header policy found."
- Agent config: partial — "https://cerebrium.ai/docs/skill.md — One official installable skill surface; no second independent agent configuration surface established."

## Compliance & regions

- SOC 2 Type II: certified (scope: hosted) — "Cerebrium has successfully completed its SOC 2 Type 2 audit." [source](https://cerebrium.ai/blog/cerebrium-achieves-soc-2-type-ii-compliance-for-secure-production-ai-infrastructure)
- ISO/IEC 27001: certified (scope: hosted) — "(2026) ISO 27001:2022 Certificate" [source](https://trust.cerebrium.ai/)
- HIPAA: self-attested (scope: hosted) — "Cerebrium offers a standardized BAA to all customers who require HIPAA compliance." [source](https://cerebrium.ai/docs/security.md)
- GDPR: self-attested (scope: hosted) — "Cerebrium supports customer obligations under the General Data Protection Regulation (GDPR)." [source](https://cerebrium.ai/docs/security.md)
- Regions: us-east-1 (N. Virginia), us-central1 (Kansas City (Nebius)), eu-north-1 (Stockholm), eu-north1 (Finland (Nebius))
- Regions note: Multi-region beta. Additional regions on request: us-west-2, eu-west-2, eu-central-1, ap-south-1, ap-northeast-1, sa-east-1, ca-central-1, me-central-1. Region-local storage is not synchronized; global apps mount a shared global volume.

## Company & presence

- Headquarters: New York City, USA; founded in Cape Town, South Africa
- Founded: 2021
- Privacy policy: https://cerebrium.ai/privacy
- Terms of service: https://cerebrium.ai/terms-of-service
- GitHub: https://github.com/CerebriumAI
- X/Twitter: https://twitter.com/cerebriumai
- LinkedIn: https://www.linkedin.com/company/cerebrium
- Discord: https://discord.gg/ATj6USmeE2
- Blog: https://www.cerebrium.ai/blog
- YouTube: https://www.youtube.com/@cerebrium

## Nearby in the atlas

- [General Compute](/products/general-compute.md)
- [Agent Sandbox](/products/agent-sandbox.md)
- [Agent Substrate](/products/agent-substrate.md)
- [Agent37](/products/agent37.md)
- [AX (Agent Executor)](/products/ax.md)
- [Beam](/products/beam.md)

## Method

Facts come from the product's own public pages. Each claim carries its evidence: a quote and the page it came from. We count capabilities rather than grade products, and "Not assessed" means we have not checked, never that the product lacks it.