How do I build my first pipeline?

Type what you want, like 'a support copilot with Whisper and Qwen, tuned for latency.' RunInfra builds and optimizes the pipeline. Chat to refine it, then deploy.

Which AI models are supported?

Vetted Hugging Face models. LLM serving is fully supported end to end. Speech, embedding, vision, and image generation models are supported for optimization and benchmarking, with managed serving in staged rollout. Gated or unsupported models are flagged before you start, not after.

How does GPU kernel optimization work?

RunInfra profiles your model across GPUs, tries quantization, KV cache, serving, and kernel tweaks, and benchmarks the best tradeoff of speed, memory, and cost.

Can I deploy pipelines as APIs?

Yes. Supported pipelines deploy as REST endpoints in one click, and managed deploy unlocks at $50 of lifetime credit purchases. If something isn't deployable yet, RunInfra tells you why instead of shipping a broken endpoint.

How is this different from using closed-source APIs?

Closed APIs hide the model and the infrastructure. With RunInfra you see both, and you benchmark open models against your own latency, throughput, and cost targets. Export the stack or run it in your own cloud to own it outright. Managed hosting is the convenient option, with a free deployment kit as your exit.

Encrypted in transit and at rest, on isolated infrastructure. Your inference data never trains anything. Run an exported kit in your own cloud and your inference data stays there. Use our managed endpoints and it passes through the subprocessors listed in our privacy policy. RunInfra is SOC 2 Type II attested.

How does the balance work?

You add credits with one-time top-ups, starting at $10, and spend from one balance, in dollars. Optional auto-recharge adds credits automatically when your balance runs low. Top-ups stay valid for one year, and the same balance covers agent plans, optimization, benchmarking, deploys, and hosted inference.

RunInfraby RightNow

Dashboard Sign in Get started

Start with $10 in credits, pay for what you run.

One balance for optimization, deploys, and the agent. No subscription. Top-ups stay valid for one year.

Pay as you go

Everything you need to optimize and ship, on one balance.

No subscription

$10 minimum top-up, pay for what you run. Unlocks are lifetime purchase totals.

Managed deploy unlocked with up to 4 replicas and 2 GPUs. Add $150 more to unlock managed deploy capacity for 16 replicas and 4 GPUs.

What's included

Quantization with AWQ, GPTQ, and FP8

All standard GPUs (T4, L4, L40S, A100, H100)

Managed deploy with scale-to-zero endpoints, from $50 lifetime

OpenAI-compatible API endpoints

Deployment kit included with every optimization, your exit is free

Agent chat, plans, and benchmarking

One balance, no seats, no per-user fees

Unlimited pipelines and versioning

New accounts start with $1 free to test the inference API.

Add credits to start, $100

Enterprise

Dedicated infrastructure, compliance, and custom volume.

Custom pricing

Everything self-serve includes, plus

Self-hosted and custom-GPU deployment

Audit logs and role-based access control (RBAC)

B200 / H200 GPU access

Custom volume and contract terms

Custom SLAs up to 99.99%

SOC 2 Type II attestation

Dedicated CSM and private Slack

Request a demo

Model APIs

Pay for model tokens, not idle capacity

Price per 1M tokens

Model

Input

Cached input

Output

DeepSeek V4 Flash

Input

$0.13

Cached input

$0.01

89% hit rate

Output

$0.27

View model

Nemotron 3.5 Lightning 30B

Input

$0.05

Cached input

$0.05

Standard rate

Output

Input

Cached input

Output

Input

Cached input

Output

Input

Cached input

96% hit rate

Output

Input

Billing basis

USD, billed on input tokens

View model

Qwen3 Embedding 0.6B

Input

$0.01

Billing basis

USD, billed on input tokens

View model

Qwen3 Reranker 8B

Input

$0.05

Billing basis

USD, billed on input tokens

Input

Cached input

65% hit rate

Output

Input

Billing basis

USD, billed on input tokens

View model

Limits and modality-specific capabilities are listed on each model page.

Compare Pay as you go and Enterprise

Pay as you go is self-serve, funded by credit top-ups. Enterprise adds private infrastructure, reserved capacity, compliance, and custom terms.

Pay as you go

from a $10 top-up

Pricing modelCredit top-ups from $10, pay for what you run

Capability milestones$50: up to 4 replicas and 2 GPUs. $250: up to 16 replicas and 4 GPUs. $10,000: up to 32 replicas and 8 GPUs. Lifetime spend.

BalanceOne balance for optimization, deploys, and inference

SeatsNo per-seat fees

Optimization techniques

Standard GPUs (T4 to H100)

B200 / H200 GPUs

Managed deployUnlocks at $50 lifetime spend

Self-hosted / custom GPUIncluded (deployment kit + BYOC)

OpenAI-compatible API

Audit logs and RBAC

SOC 2 Type II

SLANo contractual SLA

SupportPriority email

Enterprise

Custom

Pricing modelCustom volume and terms

Capability milestonesIncluded under contract

BalanceOne shared balance with custom controls

SeatsNo per-seat fees

Optimization techniques

Standard GPUs (T4 to H100)

B200 / H200 GPUs

Managed deploy

Self-hosted / custom GPUIncluded, plus custom-GPU deployment

OpenAI-compatible API

Audit logs and RBAC

SOC 2 Type II

SLAUp to 99.99%

SupportDedicated CSM and private Slack

Compare Pay as you go and Enterprise Pay as you go is self-serve, funded by credit top-ups. Enterprise adds private infrastructure, reserved capacity, compliance, and custom terms.	Pay as you go from a $10 top-up	Enterprise Custom
Pricing model	Credit top-ups from $10, pay for what you run	Custom volume and terms
Capability milestones	$50: up to 4 replicas and 2 GPUs. $250: up to 16 replicas and 4 GPUs. $10,000: up to 32 replicas and 8 GPUs. Lifetime spend.	Included under contract
Balance	One balance for optimization, deploys, and inference	One shared balance with custom controls
Seats	No per-seat fees	No per-seat fees
Optimization techniques
Standard GPUs (T4 to H100)
B200 / H200 GPUs
Managed deploy	Unlocks at $50 lifetime spend
Self-hosted / custom GPU	Included (deployment kit + BYOC)	Included, plus custom-GPU deployment
OpenAI-compatible API
Audit logs and RBAC
SOC 2 Type II
SLA	No contractual SLA	Up to 99.99%
Support	Priority email	Dedicated CSM and private Slack

Credit pricing and model-serving infrastructure answer different questions. Estimate GPU fit, rent, paid idle capacity, and API crossover for a specific model and traffic shape.