Together AI API pricing
22 models · USD per 1M tokens · verified 2026-10-23
Together AI serves 22 of the models we track, 20 of them open-weights. The cheapest is DeepSeek V4 Flash 0731 at $0.14 input and $0.28 output per million tokens. For 4 of these models, Together AI is the cheapest tracked host.
// models on Together AI
| model | lab | in $/1M | out $/1M | ctx | vs best price |
|---|---|---|---|---|---|
| DeepSeek V4 Flash 0731 | DeepSeek | $0.14 | $0.28 | 1.05M | +94.4% |
| Qwen3.5 9B | Alibaba (Qwen) | $0.17 | $0.25 | 262.1K | +68.9% |
| GLM-5.3-FlashX BEST | Z.AI (Zhipu) | $0.15 | $0.50 | 1M | −59.7% |
| GPT OSS 120B | OpenAI | $0.15 | $0.60 | 131.1K | +273.4% |
| Qwen3 235B-A22B Instruct 2507 | Alibaba (Qwen) | $0.20 | $0.60 | 262.1K | +46.3% |
| DeepSeek V4.1 Flash | DeepSeek | $0.30 | $1.20 | 1.05M | +75% |
| MiniMax-M3 | MiniMax | $0.30 | $1.20 | 1M | best |
| Qwen3 Coder Next BEST | Alibaba (Qwen) | $0.50 | $1.20 | 262.1K | best |
| DeepSeek-V3.1 | DeepSeek | $0.60 | $1.70 | 163.8K | +105.9% |
| Llama 3.3 70B Turbo | Meta | $1.04 | $1.04 | 131.1K | +571% |
| Qwen3.6 Plus BEST | Alibaba (Qwen) | $0.50 | $3.00 | 1M | best |
| DeepSeek-V3 | DeepSeek | $1.25 | $1.25 | 163.8K | +170.3% |
| Qwen3.5 397B-A17B | Alibaba (Qwen) | $0.60 | $3.60 | 262.1K | best |
| GLM-5 | Z.AI (Zhipu) | $1.00 | $3.20 | 204.8K | best |
| DeepSeek V4 Pro (0813) | DeepSeek | $1.32 | $3.96 | 1.05M | +100% |
| Muse Spark Open BEST | Meta | $1.25 | $4.25 | 1.05M | best |
| GLM-5.1 | Z.AI (Zhipu) | $1.40 | $4.40 | 200K | best |
| GLM-5.2 | Z.AI (Zhipu) | $1.40 | $4.40 | 1M | best |
| GLM-5.3 | Z.AI (Zhipu) | $1.40 | $4.40 | 1M | best |
| Qwen3.7 Max | Alibaba (Qwen) | $2.50 | $7.50 | 1M | best |
| DeepSeek R1 (Turbo) | DeepSeek | $3.00 | $7.00 | 64K | +247.8% |
| Kimi K3 | Moonshot AI | $3.00 | $15.00 | 1.05M | best |
// how to read this page
Prices are Together AI's base-tier list prices per million tokens. "vs best price" compares Together AI's blended price (3 × input + output) / 4 with the lowest price we track for the same model, whether it comes from the lab's own API or another host. Open each model for its full price history and every host that serves it.