News icon

Batch Inference is now generally available on Runpod Serverless

GPU Pricing

GPU Cloud Pricing

Updated November 4, 2026

Runpod pricing depends on the GPU workload you run: Pods for dedicated GPU instances, Serverless for API inference, and Clusters for multi-node jobs.
‍


For reserved capacity and contract pricing,
speak to our enterprise sales team.
‍

Talk to Sales
Purple glow background

Pods

Thousands of GPUs across 30+ regions. Simple pricing plans for teams of all sizes, designed to scale with you.

GPU

>80GB VRAM

B300

288 GB HBM3e
251 GB RAM
32
vCPUs
$8.99/hr
288 GB HBM3e
251 GB RAM
32 vCPUs
Deploy B300

H200

141 GB VRAM
276 GB RAM
24
vCPUs
$5.29/hr
141 GB VRAM
276 GB RAM
24 vCPUs
Deploy H200

B200

180 GB VRAM
283 GB RAM
28
vCPUs
$7.99/hr
180 GB VRAM
283 GB RAM
28 vCPUs
Deploy B200

RTX Pro 6000

96 GB VRAM
188 GB RAM
16
vCPUs
$2.49/hr
96 GB VRAM
188 GB RAM
16 vCPUs
Deploy RTX Pro 6000

H100 NVL

94 GB VRAM
94 GB RAM
16
vCPUs
$3.19/hr
94 GB VRAM
94 GB RAM
16 vCPUs
Deploy H100 NVL
80GB VRAM

H100 PCIe

80 GB VRAM
188 GB RAM
16
vCPUs
$2.89/hr
80 GB VRAM
188 GB RAM
16 vCPUs
Deploy H100 PCIe

H100 SXM

80 GB VRAM
125 GB RAM
20
vCPUs
$3.99/hr
80 GB VRAM
125 GB RAM
20 vCPUs
Deploy H100 SXM

A100 PCIe

80 GB VRAM
117 GB RAM
8
vCPUs
$1.79/hr
80 GB VRAM
117 GB RAM
8 vCPUs
Deploy A100 PCIe

A100 SXM

80 GB VRAM
125 GB RAM
16
vCPUs
$1.79/hr
80 GB VRAM
125 GB RAM
16 vCPUs
Deploy A100 SXM
48GB VRAM

Pro 6000 MIG 48GB

48 GB VRAM
62 GB RAM
8
vCPUs
$1.29/hr
48 GB VRAM
62 GB RAM
8 vCPUs
Deploy Pro 6000 MIG 48GB

L40S

48 GB VRAM
94 GB RAM
16
vCPUs
$1.09/hr
48 GB VRAM
94 GB RAM
16 vCPUs
Deploy L40S

RTX 6000 Ada

48 GB VRAM
167 GB RAM
10
vCPUs
$0.99/hr
48 GB VRAM
167 GB RAM
10 vCPUs
Deploy RTX 6000 Ada

A40

48 GB VRAM
50 GB RAM
9
vCPUs
$0.59/hr
48 GB VRAM
50 GB RAM
9 vCPUs
Deploy A40

L40

48 GB VRAM
94 GB RAM
8
vCPUs
$0.82/hr
48 GB VRAM
94 GB RAM
8 vCPUs
Deploy L40

RTX A6000

48 GB VRAM
50 GB RAM
9
vCPUs
$0.59/hr
48 GB VRAM
50 GB RAM
9 vCPUs
Deploy RTX A6000
32GB VRAM

RTX 5090

32 GB VRAM
35 GB RAM
9
vCPUs
$1.19/hr
32 GB VRAM
35 GB RAM
9 vCPUs
Deploy RTX 5090
24GB VRAM

Pro 6000 MIG 24GB

24 GB VRAM
31 GB RAM
4
vCPUs
$0.69/hr
24 GB VRAM
31 GB RAM
4 vCPUs
Deploy Pro 6000 MIG 24GB

L4

24 GB VRAM
50 GB RAM
12
vCPUs
$0.59/hr
24 GB VRAM
50 GB RAM
12 vCPUs
Deploy L4

RTX 3090

24 GB VRAM
125 GB RAM
16
vCPUs
$0.5/hr
24 GB VRAM
125 GB RAM
16 vCPUs
Deploy RTX 3090

RTX 4090

24 GB VRAM
41 GB RAM
6
vCPUs
$0.89/hr
24 GB VRAM
41 GB RAM
6 vCPUs
Deploy RTX 4090

RTX A5000

24 GB VRAM
25 GB RAM
9
vCPUs
$0.27/hr
24 GB VRAM
25 GB RAM
9 vCPUs
Deploy RTX A5000
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Serverless

Cost effective for every inference workload.

GPU

Workers
288
GB
B300

Maximum throughput for big models.

$
11.99
/hr
180
GB
B200

Maximum throughput for big models.

$
10.79
/hr
141
GB
H200

Extreme throughput for big models.

$
6.99
/hr
96
GB
RTX 6000 Pro
PRO

High throughput for large model inference workloads.

$
3.49
/hr
80
GB
H100
PRO

Extreme throughput for big models.

$
5.39
/hr
80
GB
A100

High throughput GPU, yet still very cost-effective.

$
2.72
/hr
48
GB
L40, L40S, 6000 Ada, MIG 48GB
PRO

Extreme inference throughput on LLMs like Llama 3 7B.

$
1.75
/hr
48
GB
A6000, A40

A cost-effective option for running big models.

$
1.22
/hr
32
GB
5090
PRO

Extreme throughput for small-to-medium models.

$
1.79
/hr
32
GB
RTX PRO 4500 Blackwell

Cost-effective Blackwell inference for 32GB workloads.

$
1.15
/hr
24
GB
4090
PRO

Extreme throughput for small-to-medium models.

$
1.34
/hr
24
GB
L4, A5000, 3090, MIG 24GB

Great for small-to-medium sized inference workloads.

$
0.75
/hr
16
GB
A4000, A4500, RTX 4000, RTX 2000

The most cost-effective for small models.

$
0.58
/hr
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Clusters

Launch multi-GPU clusters in minutes with no commitments. Scale up to 64 GPUs, attach shared storage, and pay only for what you use.

GPU

H200 SXM
$
6.35
/hr
H200 SXM
A100 SXM
$
2.15
/hr
A100 SXM
L40S
Contact sales
L40S
H100 SXM
Contact sales
H100 SXM
B200
Contact sales
B200
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Reserved Clusters

Dedicated GPU clusters with guaranteed availability, custom configurations, SLA-backed uptime, and discounted rates for enterprises scaling to 10,000+ GPUs.

GPU

H200 SXM
Contact sales
H200 SXM
A100 SXM
Contact sales
A100 SXM
L40S
Contact sales
L40S
H100 SXM
Contact sales
H100 SXM
B200
Contact sales
B200
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Storage

Flexible and persistent storage options starting at $0.05/GB/mo with standard and high-performance tiers.

Storage Type

Container Disk
$0.10/GB/mo
Volume Disk
Running - $0.10/GB/mo
Idle - $0.20/GB/mo
Volume Disk
Network Storage (Standard)
Under 1TB - $0.07/GB/mo
Over 1TB - $0.05/GB/mo
Network Storage (Standard)
Network Storage (High-Performance)
$0.14/GB/mo
Network Storage (High-Performance)
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Public Endpoints

Instant access to pre-deployed AI models via API. No infrastructure setup required.

Model Name

Audio
Pruna / Whisper V3 Large
$0.05 per 1000 characters
Pruna / Whisper V3 Large
resembleai / Chatterbox Turbo
$0.00 per 1000 characters.
resembleai / Chatterbox Turbo
minimax / Minimax Speech 02 HD
$0.05 per 1000 characters
minimax / Minimax Speech 02 HD
minimax / Minimax Speech 02 HD
$0.05 per 1000 characters
minimax / Minimax Speech 02 HD
Image
bytedance / Seedream 4.0 Edit
$0.0270 per request
bytedance / Seedream 4.0 Edit
bytedance / Seedream 4.0 T2I
$0.0270 per request
bytedance / Seedream 4.0 T2I
google / Nano Banana Edit
$0.0380 per request
google / Nano Banana Edit
google / Nano Banana Pro Edit
$0.14 per request
google / Nano Banana Pro Edit
pruna / Pruna Image T2I
$0.0050 per request
pruna / Pruna Image T2I
pruna / Pruna Image Edit
$0.01 per request
pruna / Pruna Image Edit
alibaba / WAN 2.6 T2I
$0.03 per request
alibaba / WAN 2.6 T2I
qwen / Qwen Image Edit 2511
$0.02 per request
qwen / Qwen Image Edit 2511
qwen / Qwen Image Edit 2511 LoRA
$0.025 per request
qwen / Qwen Image Edit 2511 LoRA
Tongyi-MAI / Z Image Turbo
$0.0050 per request.
Tongyi-MAI / Z Image Turbo
black-forest-labs / FLUX.1 [dev]
$0.02 per megapixel
black-forest-labs / FLUX.1 [dev]
black-forest-labs / FLUX.1 Kontext [dev]
$0.0250 per request
black-forest-labs / FLUX.1 Kontext [dev]
black-forest-labs / FLUX.1 Schnell
$0.0024 per megapixel
black-forest-labs / FLUX.1 Schnell
Bytedance / Seedream 3.0
$0.0300 per request
Bytedance / Seedream 3.0
qwen / Qwen Image Edit
$0.0200 per request
qwen / Qwen Image Edit
qwen / Qwen Image
$0.0200 per request
qwen / Qwen Image
qwen / Qwen Image LoRA
$0.0250 per request
qwen / Qwen Image LoRA
Language
deep-cogito / Deep Cogito v2 Llama 70B
$0.00001 per 1m tokens
deep-cogito / Deep Cogito v2 Llama 70B
qwen / Qwen3 32B AWQ
$10.00 per 1m tokens
qwen / Qwen3 32B AWQ
minimax / Minimax Speech 02 HD
$0.05 per 1000 characters
minimax / Minimax Speech 02 HD
minimax / Minimax Speech 02 HD
$0.05 per 1000 characters
minimax / Minimax Speech 02 HD
ibm / IBM Granite 4.0 H Small
$1.00 per 1m tokens
ibm / IBM Granite 4.0 H Small
Video
Bytedance / Seedance 1.0 pro
5s: $0.12(480p) per request
Bytedance / Seedance 1.0 pro
Alibaba / Wan 2.2 I2V 720p
5s: $0.30 per request
Alibaba / Wan 2.2 I2V 720p
Alibaba / Wan 2.2 T2V 720p
5s: $0.30 per request
Alibaba / Wan 2.2 T2V 720p
Alibaba / Wan 2.1 I2V 720p
$0.30 per request
Alibaba / Wan 2.1 I2V 720p
Alibaba / Wan 2.1 T2V 720p
$0.30 per request
Alibaba / Wan 2.1 T2V 720p
kwaivgi / Kling v2.6 Standard Motion Control
1-3s $0.21 per request
kwaivgi / Kling v2.6 Standard Motion Control
Alibaba / WAN 2.6 T2V
5s: $0.50 per request
Alibaba / WAN 2.6 T2V
bytedance / Seedance V1.5 Pro I2V
$0.024 per second
bytedance / Seedance V1.5 Pro I2V
kwaivgi / Kling Video O1 R2V
$0.112 per second
kwaivgi / Kling Video O1 R2V
Alibaba / Wan 2.6 I2V
5s: $0.50 per request
Alibaba / Wan 2.6 I2V
OpenAI / SORA 2 Pro I2V
4s $1.20 pre request
OpenAI / SORA 2 Pro I2V
MeiGen-AI / InfiniteTalk
$0.25 · 720p per request
MeiGen-AI / InfiniteTalk
OpenAI / SORA 2 I2V
4s: $0.40 per request
OpenAI / SORA 2 I2V
Alibaba / Wan 2.5 I2V
5s $0.25 pe request
Alibaba / Wan 2.5 I2V
kwaivgi / Kling v2.1 I2V Pro
5s: $0.45 per request
kwaivgi / Kling v2.1 I2V Pro
Alibaba / Wan 2.2 I2V 720p LoRA
5s: $0.35 per request
Alibaba / Wan 2.2 I2V 720p LoRA
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Runpod reservation savings illustration

Gain additional savings
with reservations.

Speak with our team about Serverless reservations and contract pricing.

How Runpod pricing works

Runpod pricing is based on the type of GPU workload you run. Pods are dedicated GPU instances for development and long-running jobs, Serverless bills inference workers based on usage, and Clusters support multi-node workloads and reserved capacity.

Storage and deployment choices affect total cost, so teams should choose the model that matches workload duration, traffic pattern, and control needs.

GPU prices per hour on Runpod

How much does an H100 PCIe cost per hour on Runpod?

Runpod rents the H100 PCIe at $2.89/hr on Secure Cloud, or from $1.99/hr on Community Cloud, billed per second with no commitment.

How much does an H100 SXM cost per hour on Runpod?

Runpod rents the H100 SXM at $3.99/hr on Secure Cloud, or from $2.69/hr on Community Cloud, billed per second with no commitment.

How much does an A100 PCIe cost per hour on Runpod?

Runpod rents the A100 PCIe at $1.79/hr on Secure Cloud, or from $1.19/hr on Community Cloud, billed per second with no commitment.

How much does an A100 SXM cost per hour on Runpod?

Runpod rents the A100 SXM at $1.79/hr on Secure Cloud, or from $1.39/hr on Community Cloud, billed per second with no commitment.

How much does an H200 cost per hour on Runpod?

Runpod rents the H200 at $5.29/hr on Secure Cloud, or from $3.59/hr on Community Cloud, billed per second with no commitment.

Enterprise Contracts

Need support for larger deployments or custom requirements? For reserved capacity and contract pricing, speak to our enterprise sales team.

Talk to sales

Build what’s next.

Build, train, and scale AI workloads on Runpod with cloud GPUs, Serverless, and Clusters.

Star field background