Includes
Pricing built to scale from weekend project
to global powerhouse
Non-Profit & Academic organizations please contact us for discounting
Explorer
For individuals, students, and hobbyists
Free Forever!
- ✓
- ∞Unlimited public repositorieswith unlimited collaborators
- ✓5 private repositoriesMaximum 3 collaborators
- ✓50gb of data storage
- ✓50gb of data transfer
Hacker
For small teams, and larger projects
$30.00 per month
$30.00 per month
- ✓Everything in Explorer +
- ∞Unlimited private repositories
- ✓100gb of data storagemore available
- ✓100gb of data transfermore available
Pro
For complex projects with larger data sets
$60.00 per month
$60.00 per month
- ✓Everything in Hacker +
- ✓500gb of data storagemore available
- ✓500gb of data transfermore available
Model pricing
All models are pay as you go. No upfront costs, no subscription required.
| Model | Inference cost | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Claude Fable 5 Long-horizon agentic reasoning, 1M context Text Token-based |
| ||||||||||||||||
GPT Image 2 High-fidelity images with near-perfect text rendering Image Per image |
| ||||||||||||||||
Seedance 2.0 - Reference to Video Reference-guided video with audio Video Per video output second |
| ||||||||||||||||
Seedream 5.0 Pro Precise image generation and editing Image Per image |
| ||||||||||||||||
Nano Banana 2 Lite Low-cost 1K image generation Image Per image |
| ||||||||||||||||
Seed Audio 1.0 Text-to-speech with voice cloning and prosody control Audio | N/A | ||||||||||||||||
Happy Horse 1.1 - Text to Video Text-to-video up to 1080p Video Per video output second |
| ||||||||||||||||
GLM 5.2 Coding and agents, 1M-token context Text Token-based |
| ||||||||||||||||
Happy Horse 1.1 - Image to Video Image-to-video up to 1080p Video Per video output second |
| ||||||||||||||||
Happy Horse 1.1 - Reference to Video Reference-to-video, up to 9 images Video Per video output second |
| ||||||||||||||||
GPU pricing
Dedicated GPUs charge per second of use and will automatically shut down after 15 minutes of inactivity.
| Instance type | GPU | vCPU | VRAM | System RAM | Price |
|---|---|---|---|---|---|
A10G | 1x A10G | 8 | 24 GiB | 32 GiB | $1.65/hr |
H100MIG | 1x H100_40GB | 13 | 40 GiB | 117 GiB | $1.95/hr |
H100 | 1x H100 | 26 | 80 GiB | 234 GiB | $4.87/hr |
H200 | 1x H200 | 8 | 137 GiB | 137 GiB | $9.98/hr |
4xH100 | 4x H100 | 104 | 320 GiB | 936 GiB | $15.00/hr |
8xH100 | 8x H100:8 | 208 | 640 GiB | 1872 GiB | $38.99/hr |
Fine-tuning pricing
Fine-tuning is billed per second of training time, after which the model can be deployed to a dedicated GPU for inference.
| Model | Pricing Method | Dedicated inference | Full fine-tune | LoRA fine-tune |
|---|---|---|---|---|
Gemma 4 E2B gemma-4-e2b-it Text | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |
Gemma 4 E4B gemma-4-e4b-it Text | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |
LTX-2.3 Pro ltx-2-3-pro Video | Time-based | 1x H200 $9.98/hr | 1x H200 $9.98/hr | 1x H200 $9.98/hr |
Qwen3.5 0.8B qwen3-5_0-8b Text | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |
Qwen3.5 2B qwen3-5_2b Text | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |
Qwen3.5 4B qwen3-5_4b Text | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |
Qwen3.5 9B qwen3-5_9b Text | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |
FLUX.2 Klein 4B black-forest-labs-flux-2-klein-4b Image | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |
FLUX.2 Klein 9B black-forest-labs-flux-2-klein-9b Image | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |
LTX-2 Pro ltx-2-19b-image-to-video Video | Time-based | 1x H100 $4.87/hr | 1x H100 $4.87/hr | 1x H100 $4.87/hr |