Prices
AI API prices compared
US dollars per million tokens from each provider’s own API. “Typical call” is the cost of 1,000 calls with 2,000 tokens in and 500 out. Updated .
| Model | Input | Cached | Output | Context | Typical call ×1,000 |
|---|---|---|---|---|---|
| qwen3.8-flash Alibaba (Qwen) · images · tools · reasoning · JSON schema | $0.15 | $0.016 | $0.47 | 992k | $0.54 |
| qwen3.8-omni-flash Alibaba (Qwen) · images · tools · reasoning · JSON schema | $0.15 | $0.016 | $0.47 | 992k | $0.54 |
| qwen3-vl-32b-instruct Alibaba (Qwen) · images · tools | $0.16 | — | $0.64 | 131k | $0.64 |
| qwen3-vl-235b-a22b-instruct Alibaba (Qwen) · images · tools | $0.40 | — | $1.60 | 131k | $1.60 |
| qwen3-vl-32b-thinking Alibaba (Qwen) · images · tools · reasoning | $0.16 | — | $2.87 | 131k | $1.75 |
| qwen3-vl-235b-a22b-thinking Alibaba (Qwen) · images · tools · reasoning | $0.40 | — | $4.00 | 131k | $2.80 |
| kimi-k2.7-code Alibaba (Qwen) · images · tools · reasoning · JSON schema | $0.95 | $0.19 | $4.00 | 229k | $3.90 |
| qwen3.8-max Alibaba (Qwen) · images · tools · reasoning · JSON schema | $2.00 | $0.25 | $6.00 | 992k | $7.00 |
Source: the open LiteLLM model price map (MIT), checked daily. Batch, long-context and regional prices differ; always check the provider’s own pricing page before you commit. Spotted a wrong price? Tell us with “Report an error” on any post.