Model Shaping
Pricing
Serverless Inference
Most teams start with serverless inference and move to dedicated endpoints at scale.
Price per 1M tokens
Batch API price
Model | Input | output |
|---|
Displayed prices refer to the lowest resolution/duration settings. Actual prices might vary.
Price per 1M tokens
Model | Input | output |
|---|
Displayed prices refer to the lowest resolution/duration settings. Actual prices might vary.
Model | Price per mp | Price per iMAGE | Default steps |
|---|
Prices include default steps shown above. Additional costs apply only when exceeding default steps. See full pricing details →
Price per video
Model | Price |
|---|
Price per 1M tokens
Model | Price |
|---|
Price per 1M tokens
Model | Price |
|---|
Price per 1M tokens
Model | Price |
|---|
Provisioned Throughput
Reserve dedicated capacity in throughput units (PTUs). Each PTU represents fixed capacity. The tokens-per-minute it delivers depends on the model and the token type.
Compute costs | Input | Cached | Output | Price |
|---|---|---|---|---|
Name | input | cached | output | rate |
Savings compare Together PTU cost against the selected commercial model's published list price ($/1M tokens) on the same traffic profile. Estimates assume continuous 24/7 provisioning (~43,800 min/mo).
Dedicated Inference
Deploy models on custom hardware with guaranteed performance and full control.
Single-tenant GPU instances with:
Hardware All prices per gpu per hour | On-demand Pay as you go | Reserved |
|---|---|---|
NVIDIA HGX H100 | $5.49 | |
NVIDIA HGX H200 | ||
NVIDIA HGX B200 | $8.99 | |
NVIDIA HGX B300 | ||
NVIDIA GB200 NVL72 | ||
NVIDIA GB300 NVL72 |
GPU Clusters
On-demand
Pay as you go GPU capacity on an hourly basis.
Hardware | Hourly |
|---|---|
NVIDIA GB200 NVL72 | — |
NVIDIA GB300 NVL72 | — |
NVIDIA HGX B200 | $8.19 |
NVIDIA HGX B300 | $9.99 |
NVIDIA HGX H100 | $3.99 |
NVIDIA HGX H200 | $5.99 |
On-demand hourly rates and reserved capacity
All prices are per GPU per hour.
Hardware | Preemptible Compute | ON-Demand | Reserved | |||
|---|---|---|---|---|---|---|
Pay as you go | Pay as you go | 3-30 days | 31-90 days | 91-180 days | 181+ days | |
NVIDIA HGX H100 | $1.99 | $3.99 | $3.69 | $3.45 | $3.19 | |
NVIDIA HGX H200 | $2.99 | $5.99 | $4.99 | $4.15 | $3.99 | |
NVIDIA HGX B200 | $4.09 | $8.19 | $7.99 | $7.79 | $6.79 | |
NVIDIA GB200 NVL72 | — | — | ||||
NVIDIA GB300 NVL72 | — | — | ||||
NVIDIA HGX B300 | $4.99 | $9.99 | ||||
Sandbox
Code Sandbox
Customize a deployment of VM sandboxes for large development environments.
Compute costs | Price/Hour |
|---|---|
Per vCPU | $0.0446 |
Per GiB RAM | $0.0149 |
Code Interpreter
Execute LLM-generated code securely using our API.
Duration? | Price/Session |
|---|---|
Session (60 minutes) | $0.03 |
Storage
High-bandwidth, parallel filesystem colocated with your compute.
Compute costs | Price | Unit |
|---|---|---|
Shared Filesystem | $0.16 | GiB/month |
Fine-Tuning
Train open-source models for real production use.
Per 1M tokens
Model | Supervised Fine-Tuning | Direct Preference Optimization | Minimum charge |
|---|---|---|---|
Qwen3.5 0.8B | $0.34 | $0.84 | $4.00 |
Qwen3.5 122B A10B | $4.20 | $10.50 | $10.00 |
Qwen3.5 27B | $1.05 | $2.62 | $4.00 |
Qwen3.5 2B | $0.34 | $0.84 | $4.00 |
Qwen3.5 35B A3B | $1.05 | $2.62 | $4.00 |
Qwen3.5 397B A17B | $5.60 | $14.00 | $22.00 |
Qwen3.5 4B | $0.34 | $0.84 | $4.00 |
Qwen3.5 9B | $0.34 | $0.84 | $4.00 |
Qwen3.6 27B | $1.05 | $2.62 | $4.00 |
Qwen3.6 35B A3B | $1.05 | $2.62 | $4.00 |
Qwen3.8 27B | $1.05 | $2.62 | $4.00 |
DeepSeek-V3.1 | $7.00 | $17.50 | $20.00 |
DeepSeek V4 Flash 0731 | $6.00 | $15.00 | $12.00 |
Gemma 4 26B A4B | $1.05 | $2.62 | $4.00 |
Gemma 4 31B | $1.05 | $2.62 | $4.00 |
Gemma 4 31B VLM | $1.05 | $2.62 | $4.00 |
Llama 3.3 70B | $2.03 | $5.08 | $4.00 |
Llama 4 Maverick | $5.60 | $14.00 | $16.00 |
Llama 4 Maverick VLM | $5.60 | $14.00 | $16.00 |
Llama 4 Scout | $2.10 | $5.25 | $6.00 |
Llama 4 Scout VLM | $2.10 | $5.25 | $6.00 |
Llama 3.1 8B | $0.34 | $0.84 | $4.00 |
Mixtral 8x7B Instruct v0.1 | $1.05 | $2.62 | $4.00 |
Kimi K2.6 | $15.00 | $37.50 | $60.00 |
Kimi K2.7 Code | $15.00 | $37.50 | $60.00 |
gpt-oss-120B | $2.50 | $6.25 | $6.00 |
gpt-oss-20B | $0.40 | $1.00 | $4.00 |
GLM-5.1 | $40.00 | $100.00 | $60.00 |
GLM-5.2 | $40.00 | $100.00 | $60.00 |
Price is based on the sum of tokens processed in the fine-tuning training dataset (training dataset size × number of epochs) plus any tokens in the optional evaluation dataset (validation dataset size × number of evaluations). Each job is subject to the minimum charge shown for its model.
Model | Supervised Fine-Tuning | Direct Preference Optimization | Minimum charge |
|---|---|---|---|
Qwen3.5 0.8B | $0.38 | $0.94 | $4.00 |
Qwen3.5 27B | $1.16 | $2.88 | $4.00 |
Qwen3.5 2B | $0.38 | $0.94 | $4.00 |
Qwen3.5 4B | $0.38 | $0.94 | $4.00 |
Qwen3.5 9B | $0.38 | $0.94 | $4.00 |
Qwen3.6 27B | $1.16 | $2.88 | $4.00 |
Gemma 4 31B | $1.16 | $2.88 | $4.00 |
Gemma 4 31B VLM | $1.16 | $2.88 | $4.00 |
Llama 3.3 70B | $2.24 | $5.60 | $4.00 |
Llama 3.1 8B | $0.38 | $0.94 | $4.00 |
Mixtral 8x7B Instruct v0.1 | $1.16 | $2.88 | $4.00 |
Price is based on the sum of tokens processed in the fine-tuning training dataset (training dataset size × number of epochs) plus any tokens in the optional evaluation dataset (validation dataset size × number of evaluations). Each job is subject to the minimum charge shown for its model.
No matching models
Trusted by
