Provider pricing

Together AI AI API pricing

Together AI is a leading open-source inference cloud running 200+ models on optimized H100/H200 clusters. Pricing is per-token with aggressive rates on open-weight flagships — Llama 4 Maverick, DeepSeek V4, Qwen3 Coder — plus dedicated endpoints, fine-tuning, and a serverless GPU layer for custom workloads.

Founded2022
HeadquartersSan Francisco, CA
Tracked models8

Best for

Teams running high-throughput open-source inference, fine-tuning, or needing dedicated capacity.

Strengths

  • 200+ open-source models with sub-second time-to-first-token
  • Dedicated endpoints + fine-tuning on the same API
  • Aggressive pricing on Llama, DeepSeek, Qwen, and Mixtral families
  • Together GPU Clusters for training and custom inference

Tracked Together AI models

ModelInput $/MOutput $/MContextTierBest for
Llama 4 Scout (Together AI)$0.100$0.3001,000,000BudgetHosted open-weight inference — fast, affordable general purpose
Llama 4 Maverick (Together AI)$0.200$0.6001,000,000Mid-tierHosted open-weight inference — strong reasoning and coding
Qwen3 32B (Together AI)$0.300$0.800128,000Mid-tierHosted Qwen3 32B for production agents and tool use
Mixtral 8x22B (Together AI)$1.20$3.6064,000FlagshipHosted sparse MoE — multilingual and long-context
DeepSeek V3 (Together AI)$0.900$0.900128,000BudgetHosted DeepSeek V3 with fast autoscaling inference
Llama 3.3 70B (Together AI)$0.880$0.880128,000Mid-tierHosted Llama 3.3 70B with fast turbo endpoints
DeepSeek R1 (Together AI)$3.00$7.00128,000ReasoningHosted DeepSeek R1 reasoning with autoscaling inference
Qwen 2.5 Coder 32B (Together AI)$0.800$0.800131,072Mid-tierHosted Qwen 2.5 Coder 32B for IDE-grade completions