About
Together AI is a cloud platform for running open-source models through serverless inference, batch inference, dedicated inference, and dedicated container inference. It also includes accelerated compute with GPU clusters, sandbox environments, managed storage, and fine-tuning and evaluation tools for model shaping.
The pricing page shows usage-based pricing rather than a free plan: serverless inference is priced per 1M tokens, dedicated inference starts at $3.99 per hour for 1x H100 80GB, GPU clusters start at $3.49 per hour, sandbox compute is billed per vCPU and GiB RAM, managed storage is $0.16 per GiB/month, and fine-tuning is priced per 1M tokens with model-specific minimum charges.
- Serverless inference APIs
- Batch inference workloads
- Dedicated GPU endpoints
- GPU clusters on demand
- Sandbox development environments
- Managed storage for model data
- Fine-tuning and evaluations
Free Tier Value
This pricing page does not show any concrete free tier, free credit, or trial, so there is no defensible monthly value to assign. The cheapest visible paid offering is usage-based serverless inference, but without a stated free quota there is nothing to compare against, and a credit card requirement is not explicitly waived.
What's included in the free tier
Verified
Details aren't itemized for this entry yet - check the pricing page below for the latest.
See Together AI pricing for current limits.
Paid plans
Serverless Inference
- input tokens
- per 1M tokens
- output tokens
- per 1M tokens
- Chat
- Vision
- Image
- Audio
- Video
- Transcribe
- Embeddings
- Rerank
GPU Clusters
- Accelerated compute
- Reliable GPU clusters at scale
- GB300
- GB200
- B200
- H200
- H100
Sandbox
- Developer environments
- Build development environments for AI
Managed Storage
- Storage
- Store model weights & data securely
Fine-Tuning
- Shape models with your data
- Model shaping
Provisioned Throughput
- tokens
- capacity-based
- Token-based capacity
- SLAs
- Dedicated capacity
Pricing extracted from Together AI's pricing page. Always verify current pricing before committing.
